4. EXPERIMENTAL REPORTS
4.1. CONDENSED MATTER PHYSICS
Neutron Research of the High- and Low-Temperature Phases in Rb2KfeF6 Elpasolite
A.V.Belushkin, A.I.Beskrovnyi, S.G.Vasilovsky, V.V.Sikolenko, K.S.Aleksandrov, I.N.Flerov,
Structural Study of the Lithium Orthogermanates in the Superionic State
S.N.Bushmeleva, A.M.Balagurov, R.V.Shpanchenko, V.I.Voronin, G.Sh.Shekhtman, D.V.Sheptyakov
Atomic and Magnetic Structures of Sr2GaMnO5.41 and Sr2GaMn(O,F)6
V.Pomjakushin, D.Sheptyakov, P.Fischer, A.Balagurov, A.Abakumov, M.Alekseeva, M.Rozova,
High Pressure Effects on the Crystal and Magnetic Structure of Pr1-xSrxMnO3 Manganites
D.P.Kozlenko, V.P.Glazkov, Z.Jirak, B.N.Savenko
Residual Strain Investigations Using Neutron-TOF-Diffraction on Marble Building Stone
Ch.Scheffzuek, S.Siegesmund, A.Koch, A.Frischbutter, K.Walther
Seismic Properties and Anisotropy of Rock Samples from the Kola Superdeep Well Based on
Neutron Diffraction and Seismic Velocity Laboratory Measurements
T.I.Ivankina, A.N.Nikitin, H.M.Kern
Changes in Mitochondrial Structure Induced by Swelling
T.N.Murugova, V.I.Gordeliy, A.Kh.Islamov, A.I.Kuklin, A.Nuernberg, L.S.Yaguzhinsky
Small-Angle Neutron Scattering Study of Structural Change of the Coal Tar Pitch Additived with
Nanocarbon under Heat Treatment
I.Ion, M.V.Avdeev, A.Kuklin, Y.Kovalev, M.Balasoiu, A.M.Bondara, C.Banciu, I.Pasuk
On the Possibility of Cluster Formation in Molecular Solutions of Fullerenes
T.V.Tropin, M.V.Avdeev, V.B.Priezzhev, J.W.P.Schmelzer, V.L.Aksenov
Neutron Scattering Studies of Methyl Derivatives of Benzene Selected as Potential Materials for
Cold Neutron Moderators
I.Natkaniec, K.Holderna-Natkaniec, J.Kalus
Partial Structure Features of Pb-K Melt
N.M.Blagoveshchenski, V.A.Morozov, A.G.Novikov, V.V.Savostin, A.L.Shimkevich, I.Yu.Shimkevich
First Physical Results from REMUR
V.L.Aksenov, K.N.Zhernenkov, Yu.V.Nikitenko, A.V.Petrenko
4.2. NEUTRON NUCLEAR PHYSICS
Violation of Fundamental Symmetries
Nature of the Parity Violation in Interaction of Neutrons with Lead
J.Andrzejewski, N.A.Gundorin, I.L.Karpikhin, L.Lason, G.A.Lobov, D.V.Matveev, L.B.Pikelner,
Development of Neutron Polarizer-Analyzer System for T-Invariance Experiment
V.R.Skoy, Y.Masuda, S.Muto, T.Ino, G.N.Kim
Stellar Neutron Capture of Promethium: Implications for the s-Process Neutron Density
R.Reifarth, C.Arlandini, M.Heil, F.Kappeler, P.V.Sedyshev, A.Mengoni, M.Herman, T.Rausher,
Active Biomonitoring with Moss-Bags Applied to an Industrial Site in Romania
O.A.Culicov, R.Mocanu, M.V.Frontasyeva, L.Yurukova, E.Steinnes
Neutron Activation Analysis for Development of Mercury Sorbent Based on Blue-Green Alga
L.M.Mosilishvili, A.I.Belokobylsky, A.I.Khizanishvili, M.V.Frontasyeva, E.I.Kirkesali,
Neutron research of the high- and low- temperature phases
in Rb2KFeF6 elpasolite.
A.V. Belushkin*, A. I. Beskrovnyi*, S. G. Vasilovskiy*, V. V. Sikolenko*,
K. S. Aleksandrov**, I. N. Flerov**, A..Tressaud***.
*Frank Laboratory of the Neutron Physic, JINR, Dubna, Russia
** Kirensky Institute of the Physics, Krasnoyarsk, Russia
Institut de Chimie de la Matière Condensée de Bordeaux, Pessac, France
Fluorides with general formula Rb2KМF6 belong to family of K2NaAlF6 elpasolite. Crystals
of elpasolite type at high temperatures have cubic structure and belong to sp. gr. Fm3m. At decrease
of temperature in family of crystals Rb2KМF6 there are structural phase transitions. Exception is
Rb2KAlF6 for which structural changes it was not observed, and it remains cubic down to
temperature 77К. Temperatures of transitions and sequences low temperatures phases essentially
depend on the size of a trivalent ion in M3+.
The recently neutron study of the elpasolite with large of ion radius М3+ of Rb2KBiF6 have
shown, that the crystal has two changes of crystal structure. At 360 K there is a transition in
tetragonal phase P42/mnm, and at room temperature in the monoclinic structure with sp. gr. P21/n.
At diminution of the size of ion М3+ (Tb, Ho, Y, Er) in crystals there is a trigger change of phase in
monoclinic phase Fm3m→P21/n which is connected to rotational displacements of octahedrons KF6
and MF6. In crystals with the smaller sizes of ion М3+ (Sc, In, Lu), at depression of temperature the
sequence from two transitions is observed Fm3m→I4/m→P21/n. The structure of crystals with
smaller radius of ion М3+ (М3+ = Cr, Ga, Fe) is insufficiently studied.
For investigation of the mechanism of the phase transition and definition of structure in a
low temperature phase, powder composition of Rb2KFeF6 has been synthesized and his study with
neutron diffraction is lead. The temperature of structural phase transition is equal 170 K. The
entropy change and size of temperature shift of this transition under pressure much more surpass
analogous performances for trigger transitions. Symmetry of the low-temperature phase of
compounds with small cations М3+ till now does not set, however it is supposed, that the change of
phase in them has other mechanism.
Structure of the cubic phase.
The previous investigations of Rb2KFeF6 by X-ray powder diffraction have shown  that
structure is cubic with space group Fm3m at room temperature. A cell parameter is equal 8.865 Å.
The atom were found on their special position: Rb (8c) (¼, ¼, ¼), Fe (4a) (0, 0, 0), K (4b) (½, ½,
½), F (24e) (x=0.24, 0, 0). The refinement of the crystal structure carries out with several alternative
models. In a first step of the refinement atom coordinates are fixed in they special position. In a
second step, F atoms have been distributed on rings lying in the planes orthogonal to the K-F-Fe
direction. After position refinement with anisotropic thermal vibration consideration, F has taken
position 96k. In this position F atoms are displacement to 4 equivalent states. For Rb three types of
displacement have been tested: along ,  and  directions, leading to 6, 12 and 4 local
disordered position respectively. The difference between three types of displacement is very weak
and we can not choose from it.
800 0.80 0.90 1.00 1.10 1.20 1.30
0.80 0.90 1.00 1.10 1.20 1.30
20 0 20
0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.25
d, Å d, Å
Neutron diffraction pattern of Rb2KFeF6 and the Rietveld refinement profiles at 290 K (a) and 10 K
(b). The observed diffraction data are shown at point, and the calculated profile as a solid line. Tick
marks below the profile mark the positions of allowed reflections. Difference between the observed
and the calculated intensities, normalized on a root-mean-square deviation in a point, are shown at
Structure of the orthorhombic phase
The carried out Raman scattering researches in  have shown, that candidates for space
group of the low-temperature phase are P4/m, P/m, P1121/n or P1, but they are not necessary that
only ones. Hence, low-temperature phase symmetry was unknown.
We were the carried out indexing which has shown, that position of observably diffraction
peaks it is described as tetragonal space group P4/m (a=b=6.1534(1), c=8.8939(1) Å), as
orthorhombic space group Pmnn (a=6.1567(3), b=6.1508(3), c=8.8942(3) Å). The model of atoms
position in structure for these groups was selected. The structure’s refinement was carried out. For
P4/m group the parameters of the refinement are: χ2=6.5, RW=15 %; χ2=3.2, RW=13 for Pmnn
group. A great difference of χ2 value (more than twice), allows choosing the space group for low-
temperatures phase as Pmnn.
1. A. Tressaud et al. Phys. Stat. Sol. A98, 417 (1986).
2. M. Couzi, S. Khairoun, A. Treassaud. Phys. Stat. Sol. A98, 423 (1986).
STRUCTURAL STUDY OF THE LITHIUM ORTHOGERMANATES
IN THE SUPERIONIC STATE
S.N. Bushmeleva1, A.M. Balagurov1, R.V. Shpanchenko2, V.I. Voronin3, G.Sh. Shekhtman4,
Frank Laboratory of Neutron Physics, JINR, 141980 Dubna, Russia
Department of Chemistry, Moscow State University, Moscow 119992, Russia
Institute of metal physics, Ural Branch of the Russian Academy of Sciences, 620219,
High Temperature Electrochemistry Institute, Ural Branch of the Russian Academy of
Sciences, 62021, Ekaterinburg, Russia
Laboratory for Neutron Scattering, ETHZ & PSI, CH-5232, Villigen PSI, Switzerland
The solid solution based on lithium orthogermanates in the Li4GeO4-Li3AVO4 (AV=P,V)
and Li4GeO4-Li2BVIO4 (BVI=S, Cr, Se, Mo, W) systems are considered as perspective solid
electrolytes with lithium-cation conductivity . The electric conductivity of these compositions
assumes 10-4 S*cm-1 at room
temperature while in the superionic
state region (about 870 K) it exceeds
1 S*cm-1. Li3.75Ge0.75V0.25O4 and
Li3.70Ge0.85W0.15O4 are the typical
specimens of these compounds. At
room temperature the V-containing
compound has lower conductivity than
W-based one, however at the
temperature above 500 °C the situation
is changed to opposite one (Fig.1). At
the present time there is no complete
understanding of the nature of
superionic state in these compounds.
This situation is a result of a lack of
detailed information about thermal
behavior and structural features of
these phases at high temperature.
The powder samples for
Fig. 1. Conductivity for orthogermanates. neutron diffraction investigation were
synthesized at the High Temperature
Electrochemistry Institute, Ural Branch
of the Russian academy of sciences. X-ray diffraction data at room temperature were collected at
the Department of Chemistry Moscow State University, neutron diffraction experiments – were
carried out at Frank Laboratory of Neutron Physics at IBR-2 reactor. The diffraction spectra
were measured in the heating regime at room temperature, 350 and 600 °C. The MRIA and
GSAS programs were used for structure refinement by Rietveld method. At room temperature
both diffraction patterns were indexed in the orthorhombic symmetry (S.G. Pnma). This is in
agreement with early studies of similar compounds . The neutron diffraction spectra of the
Li3.7Ge0.7V0.3O4 compound measured at room temperature is shown in Fig. 2.
Simultaneous fitting of X-ray and neutron diffraction spectra collected at room
temperature allowed to refine element distribution in studied samples. It was found that the
vanadium content is slightly higher than proposed one: about 30 % (instead of 25 %), and
lithium content decreases from 3.75 to 3.67. Thus our data revealed that the first sample has
composition Li3.67Ge0.7V0.3O4. A valence balance calculation using experimental ratio Ge/V
(0.7:0.3): w(O)V(O)+w(Ge)V(Ge)+w(V)v(V)+w(Li)V(Li)=0, where w(x) and V(x) content and
valence x element and results to lithium content of 3.7. This value is in a good agreement with
Increasing temperature up
T=25 C to 600 °C results in a noticeable
Normalized neutron count
change of the neutron diffraction
pattern. (221) and (401) peaks
coincide each with other (Fig. 3)
and all peaks on the diffraction
pattern may be indexed in
hexagonal unit cell. However both
DTA and electric conductivity
0 measurements for this sample do
1.5 2.0 2.5 3.0
not confirm a presence of the
d, Å phase transition. Except this the
Fig. 2. Experimental, calculated and difference neutron 3-fold axis is absent in the
spectra for Li3.7Ge0.7V0.3O4 at room temperature. positional relationship of the atoms
in orthorhombic unit cell (i.e. there
(221)/(401) is no hexagonal supergroups for
Pnma space group). Therefore we
concluded that at high
temperatures the Li3.7Ge0.7V0.3O4
compound has pronounced
40000 pseudosymmetry. Further
structural refinement was carried
out in Pnma space group.
(401) The lattice parameters for
o o both compounds under
T=25 C T=600 C
20000 investigation linearly increase with
rising temperature. However an
increase of the temperature does
not affect on the heavy atom
(germanium, wolfram, oxygen)
positions inside the unit cell,
whereas coordinates of the lithium
2.36 2.40 2.44 atoms are essentially changed.
Based on results of
Fig. 3. Diffraction spectra of the Li3.7Ge0.7V0.3O4 sample structure refinement one may
measured at different temperatures. suppose that a conductivity of the
V-contained phase at room temperature for the most part is due to movement of lithium atoms
along the c-axis of the unit cell (Fig. 4). In the Li3.7Ge0.82W0.15O4 structure lithium atoms are
uniformly distributed inside the unit cell and preferred direction is absent.
The difference in the transfer mechanism can explain a higher electric conductivity in
Li3.7Ge0.82W0.15O4 at the room temperature in comparison with that for Li3.7Ge0.7V0.3O4. At
higher temperatures lithium atoms in both structures are lined up along the <010> direction of
the unit cell (Fig. 5). This results in a close values of electric conductivity (Fig. 1). However in
the Li3.7Ge0.7V0.3O4 structure lithium atoms are less ordered that is probably a reason for higher
Fig. 4. Crystal structure of Fig. 5. Crystal structure of
Li0.7Ge0.7V0.3O4 at room temperature. Li0.7Ge0.7V0.3O4 at 600 °С.
1. E.I. Burmakin. Solid electrolytes with conductivity on the cations of the alkaline metals.
Nauka, М. (1992), 263 с.
2. E.I. Burmakin, G.K. Stepanov, S.V. Zhidovinova, Electrochemistry 18, (1982), 649
ATOMIC AND MAGNETIC STRUCTURES OF Sr2GaMnO5.41 AND
V. Pomjakushina,b, D. Sheptyakova, P. Fischera, A. Balagurovb, A. Abakumovc, M. Alekseevac, M.
Rozovac, E. Antipovc
Laboratory of Neutron Scattering, ETHZ & PSI, CH-5232 Villigen PSI, Switzerland
Frank Laboratory of Neutron Physics, JINR, 141980, Dubna, Russia
Department of Chemistry, Moscow State University, Moscow 119899, Russia
Layered complex manganese oxides, A2GaMnO5+x (A=Ca, Sr) with the brownmillerite-
type structure, represent a new family of manganites with possible CMR effect. These compounds
contain single MnO2 layers separated by 3 cation-oxygen layers (AO)(GaO)(AO) rather than the
more usual 2, as in the Ruddlesden-Popper system. The oxygen content can be adjusted in the
range from 5.0 to 5.5, corresponding to a nominal Mn oxidation state between +3 and +4,
respectively. For the end members, the magnetic moments of Mn are coupled
antiferromagnetically (AFM) in the MnO2 plane, but either anti- (G-type, TNG=180 K) or
ferromagnetically (C-type, TNC=100 K) between planes for x=0 or x=0.5, respectively. The change
in the magnetic ordering type from G to C is driven by the strong diagonal 180° superexchange
AFM interaction between Mn4+-ions (t32g) in adjacent layers through additional oxygen atoms in
the GaO1+x layer (coordinating Ga-ions octahedrally). In the oxidized composition (x=0.5) there is
a contribution of short-range G-type antiferromagnetic correlations, which develops below TNG and
which is partially suppressed below the transition to the AFM state at TNC. A detailed report on the
end-member properties is given in .
The intermediate valence of Mn3+2x can lead to the activation of double exchange,
producing a FM metallic state. However, the oxygen index x cannot be continuously varied from 0
to 0.5 in single-phase compositions, owing to the presence of a miscibility gap. One of the
intermediate Mn-valence single-phase compositions, which we were able to synthesize has x=0.41,
which corresponds to Mn+3.82. Its crystal structure can be satisfactorily refined in the Ammm SG
with lattice parameters a=8.00, b=5.40, c=5.36 Å at T=300 K. In this case, the (8p)-oxygen
position in the GaO1+x layer is partially filled and disordered. This disorder implies a statistical
distribution of GaO4-tetrahedra and GaO6-octahedra. An important feature of the diffraction
pattern is an anisotropic broadening of the Bragg peaks along the  direction, which is clearly
seen in high resolution neutron-diffraction data. This broadening is well modelled by anisotropic
micro strains along . The reason for the strains is probably a distribution of octahedra and
tetrahedra in Ga-layers, since the Mn-Mn distance along the a axis is larger for Mn3+, which
corresponds to the local tetrahedral Ga coordination.
The magnetic structure has a G-type AF-component below 140 K, and a C-type one below
110 K. This behaviour is similar to the one observed for x=0.5, but the G-type correlations for
x=0.41 are long-ranged. Whether these G- and C-type magnetic structures are spatially separated
remains open question.
Three fluorinated compounds Sr2GaMnO5-xF1+x with x>0 were prepared. A nominal
oxidation state of manganese determined by iodometric titration for one of the samples amounted
to +3.8 (x=0.2). The low temperature crystal structure (Fig. 1) has an orthorhombic SG Pmmm
(a<c<b, y-axis is perpendicular to the MnO2-planes), however for one of the samples the structure
can be also represented as a pseudo-tetragonal similar to the oxygenated x=0.5 composition
Fig. 1. Crystal structure of Sr2GaMnO5F1. The Fig. 2. Spin configuration in Sr2GaMn(O,F)6.
MnO2 planes and Ga(O,F)6 octahedra are shown. Only Mn-ions are shown.
(P4/mmm). The longest Mn-O distance is along z-axis, implying that the eg-orbitals of Mn3+ are
directed along z. The Mn-moments are AFM-ordered below TN≈70 K with µ(15 K)=1.5(1) µB.
Mn-spins are FM-coupled between planes, as expected for “diagonal” superexchange through
completely filled Ga(O,F)2 layers , but in-plane the Mn-spins form FM-rods along z-axis
coupled AFM to each other along x-axis, as shown in Fig. 2. This type of magnetic structure can be
well understood in frame of standard superexchange theory. The z2-orbitals are orientation
ordered, but there is no translational ordering of the orbitals due to the random distribution of the
Mn3+. Having this orientation of the orbitals the suprexchange interaction along x-direction is
always AFM, while along the z-direction it can be both FM and AFM giving the average structure,
which is shown in Fig. 2.
 V. Pomjakushin, A. Balagurov, T. Elzhov, D. Sheptyakov, P. Fischer, D. Khomskii, V.
Yushankhai, A. Abakumov, M. Rozova, E. Antipov, M. Lobanov, S.Billinge, Phys. Rev. B 66,
HIGH PRESSURE EFFECTS ON THE CRYSTAL AND MAGNETIC
STRUCTURE OF Pr1-xSrxMnO3 MANGANITES (x = 0.5 - 0.56)
D. P. Kozlenko1, V. P. Glazkov2, Z. Jirák3 and B. N. Savenko1
Frank Laboratory of Neutron Physics, JINR, 141980 Dubna Moscow Reg., Russia
Russian Research Center "Kurchatov Institute", 123182 Moscow, Russia
Institute of Physics, Cukrovarnická 10, 162 53 Prague 6, Czech Republic
Manganites of perovskite type A1-xA’xMnO3 (A - rare earth, A’ - alkali earth elements)
exhibit rich magnetic and electronic phase diagrams depending on the A (A’) - site elements
and show an extreme sensitivity of magnetic, structural, electronic and transport properties to
variation of temperature, external high pressures and magnetic fields . These systems have
attracted considerable interest with respect to the recently discovered colossal
magnetoresistance (CMR) effect.
The crystal and magnetic structures of the manganites Pr1-xSrxMnO3 (x = 0.5, 0.56)
have been studied by means of powder neutron diffraction at high external pressures up to 4.8
GPa at the DN-12 diffractometer. At ambient pressure, both compounds have a tetragonal
structure (sp. gr. I4/mcm). At TN ≈ 215 K in Pr0.44Sr0.56MnO3 a onset of A-type
antiferromagnetic (AFM) state accompanied by the structural phase transformation from the
tetragonal to the orthorhombic structure with Fmmm symmetry occurs. Pr0.5Sr0.5MnO3
exhibits an intermediate ferromagnetic (FM) state with a tetragonal structure at T < TC = 265
K and transforms to the A-type AFM state with the orthorhombic Fmmm structure at TN ≈ 175
K. Under high pressure, in originally pure A-type AFM Pr0.44Sr0.56MnO3 a phase separated
state is formed, consisting of the mixture of the A-type AFM phase with the orthorhombic
Fmmm structure (TN ≈ 220 K) and C-type AFM phase with the tetragonal I4/mcm structure
(TN ≈ 125 K). In Pr0.5Sr0.5MnO3 the application of high pressure leads to the noticeable
increase of transition temperature from FM to the A-type AFM state up to TN ≈ 230 K and the
formation of the phase separated state below ≈ 150 K, consisting of the mixture of the A-type
AFM phase with the orthorhombic Fmmm structure and the tetragonal I4/mcm phase without
long range magnetic order. Anisotropy of the lattice compression leads to the marked apical
elongation of the MnO6 octahedra in the tetragonal phase and creates favorable conditions for
appearance of the d(3z2-r2) orbital polarization, prerequisite for the C-type AFM order.
15000 P=0 GPa, A-type AFM
T=16 K Fig. 1. Neutron diffraction
patterns of Pr0.44Sr0.56MnO3
Intensity, arb. units
measured at P = 0 and 1.9
10000 P=1.9 GPa, GPa, T = 16 K at scattering
A-type T=16 K C-type AFM
P = 0 GPa
AFM angles 2θ = 90° and 45.5°
T = 16 K
4 d ,A 6 8 (inset) and processed by the
Rietveld method. A
5000 P = 1.9 GPa C-type AFM
coexistence of the initial A-
T = 16 K type AFM orthorhombic
phase with a pressure-
induced C-type AFM
0 tetragonal phase was
1 2 3 4 5 6 observed.
 E. Dagotto, T. Hotta, and A. Moreo, Phys. Rep. 344, 1 (2001).
RESIDUAL STRAIN INVESTIGATIONS USING
NEUTRON-TOF-DIFFRACTION ON MARBLE BUILDING STONE
Christian Scheffzük 1,2, Siegfried Siegesmund 3, Andreas Koch 3, Alexander Frischbutter 1 and
Kurt Walther 1
(1) GeoForschungsZentrum Potsdam, Telegrafenberg, 14473 Potsdam, Germany
(2) Frank Laboratory of Neutron Physics, Joint Institute for Nuclear Research Dubna,
141980 Dubna, Russia
(3) Geoscience Centre of the University of Göttingen, Department of Structural Geology and
Geodynamics, Goldschmidtstr. 3, 37077 Göttingen, Germany
To describe and explain the effects of bowing on marble facade panels neutron time-of-flight
diffraction was applied for residual macro-and microstrain determination on the mineral
calcite. The results were supplemented by the determination of the crystallographic preferred
orientation (texture) of calcite by neutron diffraction as well as studies on microfabric features
of the specimens using optic microscopy.
Experimental: Microstructure, Texture and Residual strain
Durability is an important property to characterize natural rocks for exterior use. Marbles for
instance frequently show a bowing of facade panels after a short time of exposure. This
bowing is generally accompanied with a reduction of strength properties . For a better
understanding of the observed effect, three samples of marble (calcite CaCO3) were
investigated: a fresh broken marble (P1), a good conditioned facade panel (P2) and a strong
deformed facade panel (P3). The studied samples are characterised by a wide grain size
distribution, from medium to coarse grained: the medium grain size is between 1 and 2 mm
with a maximum of up to 6 mm. Domains with a coarser grain size exhibit a polygonal to
interlobate shape, and straight to slightly curved grain boundaries (Fig. 1a). Evidence of
crystal-plastic deformation is documented by deformation twins and undulatory extinction.
Furthermore, the fabric is characterised by a preferred grain boundary orientation more or less
parallel to the foliation (Fig. 1b).
Fig. 1: Thin section images from demounted panels: (a) section vertical to the foliation (weak
bowing of a good conditioned panel P2); (b) section vertical to the foliation (strong bowing of
a strong deformed panel P3): the open grain boundaries are clearly visible (arrows).
In thin sections from strongly bowed panels, open grain boundaries, which are connected to
intergranular microcracks, can be observed. The observed cracks are opened up to 0.5 mm
with a length up to 5 mm (Fig. 1b). Intracrystalline cracks along twin planes are more rare.
Apparently there is a correlation between the presence of microcracks and bowing. In contrast
to the strong deformed sample, the fresh broken, undeformed sample does not show any
evidence of open grain boundaries. For P2 a warping of 0.2 mm/m was measured while for
sample P3 17.1 mm/m was observed.
The fresh broken Peccia marble sample (P1) was measured using neutron time-of-flight
diffraction at the texture diffractometer SKAT. The sample exhibits a strong crystallographic
preferred orientation. The (0006) pole figure of calcite shows an maximum normal to the
macroscopic foliation with a weak tendency to form a girdle_ distribution around a maximum
of the a-axis distribution in the foliation plane, while the (11 2 0) poles are arranged on a great
circle around the (0006) pole maximum (Fig. 2). The crystallographic a-axes corresponding to
the (11 2 0) poles are oriented within the foliation plane. The importance of calcite textures to
the contribution of physical weathering has been widely discussed. A general observation is
that the maximum deterioration is closely linked to the c-axis maximum. Only the texture of
the fresh broken sample (P1) is shown, because the texture of the good conditioned plate (P2)
and the strong deformed facade plate (sample P3) is similar.
Fig. 2: Crystallographic preferred orientation (texture) of the fresh broken Peccia marble (P1),
measured by neutron-TOF-diffraction at SKAT, projection into the foliation plane.
The strain measurements were carried out at the diffractometer EPSILON-MDS at beam line
_ _ _ _
7A _[2, 3]. Six Bragg reflections of calcite (01 12), (10 14), (0006), (11 2 0), (11 2 3), and
(01 18) were investigated. Macroscopic internal strains (g = )d/d) at all samples in relation to
the stress free state were determined by analysing the position of the Bragg peaks (Fig. 7).
The stress free state as the reference value were determined by measuring rock powder,
prepared by grinding up and annealing. Microscopic internal stresses, caused by dislocations
and other microscopic defects, could be observed by peak broadening (Fig. 7). The figures
show the dependence of macroscopic and microscopic strain in dependence on the detected
six Bragg reflections.
Macro- and microstrain data for the acquired direction perpendicular to the foliation plane are
shown in Figure 3. The (01 12) Bragg reflections for all samples are characterized by a
positive strain. The microscopic strain shows no significant differences between the three
investigated samples. Only the good conditioned and the strong deformed facade panel show
positive strain at the (10 14)-Bragg reflection, whereas the fresh broken facade panel shows a
negative strain. The good conditioned sample shows a positive strain of g = +(720±150)x10-6,
the strong deformed facade panel a lower tensional strain of g = +(380±140)x10-6. All three
panels show comparable macroscopic positive strain values for the c-axis . The
microscopic strain is characterized by a lower FWHM for the fresh broken sample in contrast_
to a little larger FWHM for good conditioned and the strong deformed sample. The a-[11 2 0]-
axes are characterized by the highest positive strain value at the strong deformed facade panel
with g = +(980±170)x10-6, the fresh broken sample shows a tensional strain of
g = (200±140)x10-6, whereas the good conditioned facade panel shows a negative
compressive strain of g = -(420±180)x10-6.
Fig. 3: Macrostresses (left) and microstresses (right), measured by neutron time-of-flight
diffraction at EPSILON-MDS, perpendicular to the foliation plane.
In the acquired direction parallel to the foliation plane, also all three samples show
comparable macroscopic strain values for the c--axis, the a-[11 2 0]-axis, the lattice
planes (11 2 3) and (01 18), but with a tendency to compression in the strong deformed facade
panel. Microscopic strain by peak broadening were found at the strong deformed facade panel
(P3) on the a-[11 2 0]-axis with a full width at half maximum (FWHM) of (21.8 ± 1.0) time
channels*32 µs in relation to a FWHM of (17.8 ± 0.3) time channels * 32 µs for the fresh
broken sample and the good conditioned facade panel.
The measured residual strain values acquired perpendicular to the foliation plane are higher
than in the direction parallel to the foliation plane. A dependence on the bowing process of the
plates may be concluded, because the samples are mainly bowed either concave or convex in
relation to the foliation plane. The observed texture is mainly characterized by a preferred
orientation of the basal (0006)-planes perpendicular to the foliation plane. The strong texture
of the Peccia marble is a significant evidence for plastic deformation.
This work was supported by the BMBF grants (03-DUO3X4 and 03-DU03G1).
 Koch, A. and Siegesmund, S., 2002. On site damage analysis of buildings showing
bowing of marble slabs: Fabric vs. type and degree of damage. Geol. Soc. Spec. Publ. 205:
 Frischbutter, A., Neov, D., Scheffzük, Ch., Vrána, M. and Walther, K., 2000. Lattice strain
measurements on sandstones under load using neutron diffraction. J. Struct. Geol. 22
 Walther, K., Scheffzük, C. and Frischbutter, A., 2000. Neutron time-of-flight
diffractometer Epsilon for strain measurements: layout and first results. Physica B,
Condensed Matter 276-278: 130.
SEISMIC PROPERTIES AND ANISOTROPY OF ROCK SAMPLES FROM
THE KOLA SUPERDEEP WELL BASED ON NEUTRON DIFFRACTION
AND SEISMIC VELOCITY LABORATORY MEASUREMENTS
T.I. Ivankinaa , A.N. Nikitina and H.M.Kernb
Joint Institute of Nuclear Research, Frank Laboratory for Neutron Physics,
141980, Dubna, Russia
Institute für Geowissenschaften, Universität Kiel, 24098 Kiel, Germany,
The neutron diffraction texture analysis of multiphase rocks from deep levels of the
lithosphere in a combination with traditional geophysical and geological methods is applicable
widely to study of different physical properties of rocks .
The results of experimental and theoretical investigations on two fine-to medium- grained
foliated biotite plagioclase gneisses (K8802, K9002) and two fine-grained amphibolites (K8752,
K11345) recovered from the Archean basement of the Kola superdeep well (SG-3) are presented. In
this investigation, the sample reference frame A, B, C is used which is basically related to the
borehole axis: [C] is parallel to the borehole, and [B] and [A] normal to it, with a rough relationship
of [A] to lineation. Two different methods are using to determine the seismic aisoptropy and to
discriminate between the contribution of oriented cracks and lattice preferred orientation (LPO) of
the minerals to bulk anisotropy, and to elucidate the relationship between the crystallographic fabric
and the elastic properties such as velocity anisotropy, shear wave splitting and shear wave
polarisation. First, P- and S-wave velocities in three orthogonal directions were measured as a
function of pressure and temperature, and second, 3D-velocities were calculated from measured
LPO (texture) and the known single-crystal properties. The LPO of the rock-forming minerals was
measured by TOF (Time Of Flight) neutron diffraction.
The measurements of compressional (Vp) and shear wave velocities (Vs) at pressure and
temperature were performed on oven-dried (120°C) cube-shaped specimens (43 mm edge length) in
a multi-anvil pressure apparatus using the ultrasonic pulse transmission technique with transducers
(lead zirconium titanate) operating at 2 MHz. The special arrangement of the apparatus allows
simultaneous measurements of compressional and orthogonally polarised shear wave velocities
(S1,S2) in three perpendicular directions. A detailed description of the experimental technique is
given by Kern et al..
Measurements were done over a range of pressures up to 600 MPa at room temperature and
from room temperature up to 600°C at 600 MPa confining pressure. Each set of experimentally
determined data comprises three P-wave velocities, six S-wave velocities, and the pressure (and
temperature) dependent linear (and volumetric) strain.
As an example, Figure 1 shows the directional dependencies of P-velocities for the
amphibolite sample K8752. In all samples, the velocity versus pressure relations for P-waves show
typical slopes: a steep, non-linear velocity increase up to about 200 MPa, giving way for linear
behaviour at higher pressures. Anisotropy is almost highest at low pressures resulting from a
constructive interference of the effects caused by effective oriented microcracks (shape texture) and
by lattice preferred orientation (mathematically expressed by the orientation distribution function
ODF) of the major minerals. Increasing pressure reduces the effect of cracks in a non-linear slope
approaching nearly constant values at high pressures. The pressure-dependent part of velocity
anisotropy must be attributed to oriented cracks and their progressive closure, and the residual,
almost pressure-independent part of seismic anisotropy is mainly due the ODFs of major minerals.
Fig.1. P-wave velocities and anisotropy A-Vp of a) as a function of pressure at room
temperature; b) as a function of temperature at 600 MPa confining pressure.
Neutron diffraction was applied for the determination of lattice preferred orientation (LPO)
of the major rock-forming minerals. The measurements were carried out at the texture
diffractometer SKAT of the pulsed reactor IBR-2 (Dubna, Russia), using TOF method which is, in
particular, appropriate for the investigation of materials composed of low-symmetry minerals such
as samples under the investigation, because several pole figures can be measured simultaneously.
From the experimental pole figures the orientation distribution functions (ODFs) for the
predominant mineral phases were recalculated applying the WIMV method and the computer
program BEATREX . The spatial distributions of P- and S-wave velocities for the mineral
phases were calculated from the ODFs and the corresponding single crystal elastic constants, and
the averaged bulk sample velocity distributions were determined by summarizing the velocity
distribution of the aggregates according to the volume fractions of the constituent minerals.
The calculated three-dimensional variations of Vp, Vs1 and Vs2 and shear wave splitting
Vs1-Vs2 of the prevailing rock forming minerals (hornblende, plagioclase and biotite) and the
corresponding averaged data of the bulk amphibolite sample K8752 are shown in Figure 2. The
stereoplot coordinate system corresponds to the reference frame A, B, C of the sample cube and the
experimental pole figures. In Fig.2 (right) we have rotated the diagrams along with the sample
reference frame A, B, C to bring them in accordance with the standard setting used in structural
geology: Z (top) = normal to foliation; Y (center) = parallel to foliation and normal to lineation, X
(E-W) = parallel to lineation. It is clear from the diagrams, that the Vp distribution of amphibolite
K8752 exhibits an overall rhombic symmetry. The marked Vp-anisotropy (6.81 %) calculated for
the amphibolite sample K8752 is mostly caused by the strong LPO of the anisotropic hornblende
minerals (Fig.2) and their high volume percentage. Interestingly, the Vp velocity distribution
calculated from the LPO of the constituent plagioclase minerals give hints for a dissolution of the
bulk Vp-anisotropy because it superimposes the hornblende pattern in a deconstructive way.
P-wave velocities are highest subparallel to lineation within the foliation and lowest normal
to foliation. On the Vs1-Vs2 diagram of the amphibolite sample, an overall rhombic symmetry is
also apparent. The foliated amphibolite exhibits market shear wave splitting (Vs1-Vs2) within the
foliation plane with the fast split shear wave (see on the orientation of the Vs1-polarization plane)
parallel to foliation. Normal to foliation (parallel to Z), a second shear wave will practically not be
Fig. 2. Calculated three-dimensional variations of the elastic properties of the
amphibolite sample K8752 based on the neutron diffraction measurements. The
model composition of the rock sample is displayed by the pie diagram.
The numerical calculations based on the TOF-method give important information on the
different contribution of the various rock-forming minerals to bulk elastic anisotropy and on the
relationship between the crystallographic fabric (LPO) and the seismic properties of the rocks such
as velocity anisotropy, shear wave splitting and shear wave polarisation.
 Nikitin A.N., Ivankina T.I. Physics of Particles and Nuclei. 35, № 2 (2004).
 Kern H., Liu B., Popp T.. J. Geophys. Res., 102, 3051-3065 (1997)
 Matthies S. Materials Science Forum, 408-412, 95-100 (2002).
CHANGES IN MITOCHONDRIAL STRUCTURE INDUCED BY SWELLING
T. N. Murugova*, V. I. Gordely*#&, A. Kh. Islamov*, A. I. Kuklin*, A. Nuernberg+, L. S. Yaguzhinsky+.
Frank Laboratory of Neutron Physics, Joint Institute for Nuclear Research, Dubna, Russia
A.N. Belozersky Institute of Physico-Chemical Biology, Moscow State University, Russia
IBI-2, Forschungszentrum Juelich 52425, IBI-2 Germany
Centre for Biophysics and Physical Chemistry of Supramolecular Structures, Moscow Institute for
Physics and Technology , Dolgoprudny, Russia
The mitochondria are organelles of the cell,
which produce the energy by oxidative a
phosphorylation. The oxidative phosphorylation is 10
one of the major processes providing living organisms
with energy. It has been established that swelling of
mitochondria is coupled with functional changes in 1
them. These changes are accompanied by
reorganization of mitochondrial membrane .
Earlier the influence of uncoupler on mitochondrial 0.1
structure has been investigated by method of small
angle x-ray scattering . In our experiments small 0.01
angle neutron scattering (SANS) was used to study 0.01 0.1
structural changes of mitochondria in two types of
medium - isotonic and hypotonic.
SANS experiments with intact rat liver b
mitochondria were carried out on the YUMO 10
spectrometer (reactor IBR-2, Dubna) [3, 4].
Measurements were made using the method of
contrast variation, when the mixture of various ratio 1
of H2O and D2O was in the media. The method of
contrast variation allows one to separate the scattering
curves from protein and lipidic components of 0.1
Determined scattering curves are presented in 0.01
Figure 1. In logarithmic coordinates the scattering 0.01 0.1
curves have linear sites both for initial and swelling -1
mitochondria. The slope of these sites is about –2, Figure 1 - Small angle scattering
which can be evidence of a fractal structure of the curve for mitochondria placed in
mitochondrial membrane. In the case of hypotonic isotonic (а) and hypotonic (b)
D2O-medium the scattering curve has a structural solutions.
maximum at q = 0.045 Å-1 (Fig. 1b), which
corresponds to the distance of about 140 Å in real
space. In the case of isotonic medium scattering curve
has not this particular feature (fig. 1a). The
appearance of this maximum correlates with the
adjacency of outer mitochondrial membrane to inner one and with the cristea adhesion, that have been
discovered in our previous experiments on electron microscopy .
Data of electron microscopy and small angle neutron scattering enable to calculate the distance
between membranes of adhering cristea. This distance composes approximately 65 Å. With the help of the
contrast variation method it was shown that the appearance of the maximum was determined by neutron
scattering from lipidic component of the mitochondrial membrane, while membrane proteins gather on
the membrane surface in groups. The location of the groups, in its turns, in cristea is disordered.
1. I. P. Krasinskaya, I.S. Litvinov, S.D. Zaharov, L. E. Bakeeva, L. S. Yaguzhinsky. Two qualitatively
different structural-and-functional states of mitochondria, Biochim. Т.54 №9. 1989. 1556-1561.
2. С. Mannella, D. Parsons, Uncoupler – induced changes in mitochondrial structure detected by small-
angle x-ray scattering, Biochim. et biophys. Acta, 460 (1977) 375 – 378.
3. Kuklin A.I., Islamov A.Kh., Gordely V.I. "Two Detectors System for Small Angle Neutron Scattering
Instrument", Submitted to J. Appl. Crystal.
SMALL-ANGLE NEUTRON SCATTERING STUDY OF STRUCTURAL
CHANGE OF THE COAL TAR PITCH ADDITIVED WITH
NANOCARBON UNDER HEAT TREATMENT
I.Iona,b, M. V. Avdeeva, A.Kuklina, Y. Kovaleva,
M. Balasoiua, A. M. Bondarab, C. Banciub, I. Pasukb
Joint Institute for Nuclear Research, Frank Laboratory, Dubna, Russia
Advanced Research Institute for Electrical Engineering, Bucharest, Romania
Introduction. The carbon composite materials are now one of the major structural materials
for many industrial applications due to a lot of advantages. Among them are the following: at
high-temperature the mechanical properties don’t change too much; they posses a low
density, a low coefficient of thermal expansion, as well as a good electrical conductivity, etc.
The coal tar pitch is a suitable precursor for preparation the advanced carbon material with
The sample preparation. The coal tar pitch (CTP), was processed for QI removal using a
modified Soxhlet. Characteristics of selected CTP and fractionated coal tar pitch (FCTP).
FCTP was mixed with nanocarbon powder in amounts of: 0.1%(wt). The mixtures were heat-
treated (HT) at 440oC and 900°C, with a heating rate of 1oC in controlled atmosphere (3h
soak time for each final temperature).
Results and discussions. The aim of research was to investigate the influence of additive,
nanocarbon, and high temperature (HT) on the fractionated coal tar pitch at the nanoscale by
means of small-angle neutron scattering (SANS). The nanocarbon composites materials (NC),
coke and semi-coke were investigated. SANS measurements were performed at the YuMO
spectrometer at the IBR-2 pulsed reactor, Dubna.
The linear relationship for large q-values in double logarithmic plot of the obtained scattering
curves (Fig.1) indicates to the fractal structure of material. The fractal structure has a surface
fractal dimension. A smoothness effect of the additive and HT is observed. It is quite large for
HT resulting in about 20 % increase in surface fractal dimension DS both for CTP and NC,
while the effect of the additive to DS does not exceed 2%. The smoothness effect because of
the HT treatment can be explained by the volatile releasing process during carbonization. The
HT treatment also results in an increase in the absolute scattering intensity at small q-values
for both types of the studied materials, which reflects an increase in pores and staked
arrangements of the graphene layers with the temperature. Vice verse, the incoherent
background at large q-values decreases at HT, which corresponds to a lost of some amount of
hydrogen and to an increase in crystalline order.
The appearance of the scattering nonhomogeneities at nanoscale in the studied materials is be
due to the incomplete crystallization, the absorption of amorphous mater or of same molecule
of the graphene sheets or/and the variation in size of the interlayer spacing, the diameters of
basal planes and the number of the stacked layers inside the crystalline clusters.
Conclusions: first results of the present study shows the importance of the nanoscale for the
carbon materials in their production and control over the properties. In particular, one can see
that the temperature formation of MS is higher in NC, which acts as an active site during the
formation of mesophase following an increase in the number of crystalline microdomains. In
this way new materials are more suitable for electrical purpose.
Fig.1. Obtained SANS curves.
Table 1. Results of fit of function Aq-S + B to scattering curves in double logarithmic plot
Sample/THT(oC) Slope S Fractal Parameter A Parameter B Bulk
dimension DS density
FCTP/440 3.55±0.3 2.45±0.3 1.67 0.13 1.076
NC/440 3.57±0.02 2.43±0.02 1.66 0.16 1.026
FCTP/900 3.76±0.02 2.23±0.02 2.35 0.07 1.166
NC/900 3.78±0.02 2.22±0.02 2.70 0.08 1.165
1. D.Sen, S. Mazumed, R. Chitra, J. of Materials Sciente 36 (2001) 909-912
2. P.U. Sastry, D.Sen, S. Mazumed, K.S. Chandrasekaran- Solid State Comunication 114 (
3. W. Ruland, Carbon 39 (2001) 287-324
4. X.K. Li, Zh.H. Li, D. Wu, Sh.D. Shen.-Carbon 38 (2000) 623-641
5. Rong He, Xuchaghe Xu, Changhe Chen, Hongli Fan , Bin Zhang-Fuel vol.77, no.12,
6. T.Nakagawa, I. Komaki, M.Sakawa, K. Nishikawa-Fuel 79(2000) 1341-1346
7. J.Medek, Z.Weishauptova –Fuel 79 (2000) 1621-1626
8. A.Ch. Mitropoulus, K.L. Stefanopolus, N,K. Kanellopoulos – Micrporous and
Mesoporous Materials 24(1998) 29-30
ON THE POSSIBILITY OF CLUSTER FORMATION IN
MOLECULAR SOLUTIONS OF FULLERENES
T.V.Tropin1, M.V.Avdeev1, V.B.Priezzhev2, J.W.P.Schmelzer2,3, V.L.Aksenov1
Frank Laboratory of Neutron Physics, JINR, Dubna, Russia
Bogolubov Laboratory of Theoretical Physics, JINR, Dubna, Russia
University of Rostock, Germany
First order phase transitions, condensation and cluster growth in matter play a major
role in various scientific and technological problems. The study of possible ways to describe
the time evolution of systems where aggregation processes take place, lead to the basics of the
kinetic theory. One of the modern approaches, which develop the description of such systems,
is the nucleation theory . Solutions of C60 are an example of a system, where the problem
to describe cluster growth processes arises. Fullerene clusters are detected in a number of
“C60/organic solvent” systems, as well as in triple systems “C60/organic solvent/polar
solvent”. Also, a variety of ways were developed to produce dispersions of C60 in water.
Solutions of fullerene molecules in organic non-polar solvents inhibit a variety of
unique properties, including solvatochromism and non-monotonous temperature dependence
of the solubility . The latter was assumed  to be the result of the cluster state of C60
molecules in solutions. A phenomenological theory  based on this assumption and used the
liquid droplet model of clusters is in partial agreement with experimental observations. In
particular, previous  and recent  experiments on small-angle neutron scattering from
molecular solution of C60 fullerene in carbon disulfide (CS2) do not show clusters with the
size distribution function predicted by this theory.
The aim of the present work was to use the nucleation theory approach and to check
out whether simple models for cluster growth (in particular, the liquid drop model) may result
in a cluster state of fullerenes in molecular solutions of C60.
The nucleation theory describes the formation and growth of clusters in solutions.
Using different forms for the work of cluster formation ∆G(n), a set of kinetic equations is
obtained, which describes the time evolution of the cluster size distribution function f(n, t).
c (t )
An important parameter, which determines the evolution is the ratio 0 ∞ , where c0 ( t ) is
the monomer concentration in solution and ceq is the concentration of segregating particles of
the ambient phase needed for equilibrium coexistence of both phases with a planar interface.
c (t ) c0 ( t )
The cases when 0 ∞ < 1 and ∞ > 1 correspond to homophase and heterophase
fluctuations, respectively, and should be treated separately.
The following expressions of ∆G(n) were considered:
(i) liquid droplet model: ∆G (n) = −n∆µ + α 2 n 2 / 3 ;
(ii) limited cluster growth: ∆G ( n ) = − ∆µn + α 2 n 2 / 3 + kn β ;
(iii) charged cluster model: ∆G = −n∆µ + α 2 n 2 / 3 +
1 Q2 5/3
4πεε 0 r
n −n . )
The typical thermodynamical parameters of C60 fullerene organic solutions (carbon disulfide,
benzene and toluene) were used when modeling the evolution of the f(n, t) function. One
example of such evolution obtained for model (ii) is given in Fig.1.
Fig. 1. Evolution of the cluster size distribution in time (in specific time units) obtained
numerically for model (ii) in the case of homophase fluctuations.
A special respect was given to the liquid droplet model (i). Both analytical and
numerical treatment of the corresponding kinetic equations results in the following
conclusions. In the case of homophase fluctuations a stable cluster size distribution in the
system is obtained in the form:
∆G ( n )
f ( n, t ) = Ae k BT
The maximal mean cluster size that can be obtained in the system does not exceed n ≈ 1.6 .
In the case of heterophase fluctuations the system takes a long-time evolution, the final state
being one big cluster in equilibrium with ceq monomers around it. The size of this cluster
depends on the initial supersaturation in respect with ceq (Fig.3). The obtained conclusions
again are in disagreement of the SANS from solutions C60/CS2, which shows that liquid
droplet model cannot describe the observed clusters.
The modified liquid droplet model (ii) uses the potential arising in systems where the
cluster growth is strictly limited by some kind of interaction described by the additional term
with parameters k and β. The corresponding evolution of the f(n, t) function (Fig. 1) show that
in equilibrium the final cluster size distribution consists of a Gaussian peak of large clusters
and an exponential decay distribution of monomers, dimers, trimers and so on,. The position
of Gaussian peak, as well as the average cluster size does not depend on the initial value of
c0 . The analytical expression for the mean cluster size obtained for this model:
3 β −2 α2 3 β −2
n =w s 3k (β − 1)
where ws – is the volume of a monomer (fullerene), is in agreement with numerical
solutions. This is in a qualitative agreement with the situation in the system C60/CS2, the
quantitative comparison is in progress.
First order exponent decay fit 10
0 2 4 6 8 4 6 8 10 12 14 16 18 20
α2 /kB T c0/ceq
Fig.2. Dependence of average cluster size Fig.3. Dependence of average cluster size
on system parameters in model (i) in the on initial supersaturation in model (i) in
case of homophase fluctuations. the case of heterophase fluctuations.
The last model (iii) takes into account the charge of monomers, Q, suggested for
molecular solutions in . This kind of potential can be brought to the form of ∆G(n) for
model (ii), and all the conclusions obtained above stay true for it.
1. Nucleation Theory and Applications, Eds. J.W.P.Schmelzer, G.Röpke, V.B.Priezzhev, ,
JINR, Dubna, 1999.
2. B.N.Bezmelnitstin, A.B.Eletsky, M.V.Okun, Uspekhi Physicheskih Nauk, 168 (1998)
3. Y.B.Melnichenko, et al. J. Chem. Phys. 111 (1999) 4724.
4. T.V.Tropin, M.V.Avdeev, A.A.Khokhryakov, V.B.Priezzhev, J.W.P.Schmelzer,
V.L.Aksenov, In Book of Abstracts of The XVII Workshop on the Use of Neutron
Scattering in Condensed Matter Research (RNIKS 2002), October 14 – 19, 2002,
Gatchina, PNPI RAS: Gatchina, 2002, ISBN 5-86763-061-7, p. 172.
5. A.D.Bokare, A.Patnaik, J. Chem. Phys, 119 (2003) 4529.
NEUTRON SCATTERING STUDIES OF METHYL DERIVATIVES OF
BENZENE SELECTED AS POTENTIAL MATERIALS FOR COLD
I. Natkaniec, a,b K. Holderna-Natkaniec,c J. Kalusd
Frank Laboratory of Neutron Physics, Joint Institute for Nuclear Research, 141980 Dubna, Russia
H. Niewodniczanski Institute of Nuclear Physics, 31-342 Krakow, Poland
Institute of Physics, A. Mickiewicz University, 61-614 Poznan, Poland
Institute of Physics, University of Bayreuth, D-95440 Bayreuth, Germany
Methyl derivatives of benzene, such as toluene – CH3C6H5, m-xylene – (CH3)2C6H4, and
mesitylene – (CH3)3C6H3, are commonly known as organic solvents with relatively low melting points:
180K, 225K and 227K, respectively. Melting point of p-xylene, equal to 286 K, is close to this value of
benzene, equal to 278 K. For all these molecules internal barriers for rotation of CH3 groups with regard to
C-C bound is very low. The librational modes of methyls in solid p-xylene are mixed with the high wave-
number optical phonons at the cut-off spectrum of lattice modes at about 120 cm-1 . In the solid state of
toluene  and m-xylene , methyl librations are mixed with the low wave-number acoustic phonon
branches and form the librational band at about 50 cm-1, which suggests low external barriers for rotation
of CH3 groups in the solid state of these compounds. Mesitylene was assumed to behave similarly,
therefore, it has been proposed as a cold moderator at pulsed neutron sources . However, our recent
neutron scattering investigation of this compound indicated three crystalline phases of mesitylene .
The vibrational spectra of crystalline toluene, m-xylene and mesitylene are characterized by
continuous phonon density of states with the parabolic dependence on the wave-number up to about 50
cm-1, and the phonon cut-off energy at about 120 cm-1. Except for the phase III of mesitylene, methyl
librational modes in solid toluene, m-xylene and phase I and II of mesitylene are mixed with the lattice
vibrations and increase the number of the low energy modes, which can effectively slow down neutrons to
low energies. In the disordered phase of toluene and mesitylene the bands at 50 cm-1 are smeared out and
cause a non-parabolic dependence of the G(ν) below this wave-number. Such behavior of these bands, as
well as the disappearing of the strong bands seen at phase III of mesitylene at 155 and 193 cm-1, allows
one to assign these bands as corresponding to methyl librations. Other internal modes form discrete
spectra in the range of wave-numbers from about 200 to 1800 cm-1, except for the C-H stretching modes,
which form a narrow band at about 3200 cm-1.
Fig. 1. The amplitude weighted vibrational
density of states – G(ν) obtained from the IINS
spectra measured at 20K for solid toluene, Fig. 2. The G(ν) spectra of different solid phases of mesitylene
m- and p-xylenes. at 20K.
The additional density of states, over the parabolic dependence of the G(ν) of ordered crystals,
called “boson peak”, is typical for all disordered or glassy substances. The orientational disorder of methyl
groups in solid toluene and mesitylene caused so-called protonic glass phase in solid state of these
substances. However, these glassy phases are not stable in all temperature range of solid phase of these
The disordered solid phase II of mesitylene can be obtained when overcooled liquid is freezing at
slow cooling rate. This phase is not stable at low temperatures but at about 90K it passes the structural
phase transition to the low temperature phase III. This transition is reversible until phase II is not heated
over 180K. At about 190K phase II starts a transformation to the high temperature ordered phase I. The
growth rate of nucleations of phase I increases with temperature and at about 220K full transformation of
phase II to phase I need only some minutes. This transformation is not reversible and the structure of
phase I is stable from its melting point at 227K, down to the liquid helium temperatures. Phase II, in
presence of nucleations of the phase I, can be also overcooled to the liquid helium temperatures.
Fig. 3. Comparison of the G(ν) spectra of disordered Fig. 4. Comparison of the G(ν) spectra of ordered
phases of toluene and mesitylene with its 3:2 volume crystalline phases of m-xylene and mesitylene with its
solution measured at 20K. 3:1 volume solution measured at 20K.
The IINS and neutron diffraction investigations of selected solutions have confirmed stabilization
of disordered solid phase by mixing of mesitylene with other methyl-benzene compounds. Some of these
results are presented in figures 3 and 4. It has been shown that solutions of mesitylene with toluene or m-
xylene form glassy solids, which are stable in the whole temperature range below the melting point. The
vibrational spectra of these glassy state solutions indicate that methyl librations are mixed with the lattice
vibrations and form the wide band with cut-off at about 120 cm-1.
Additional density of states at low frequencies, over the parabolic dependence of the G(ν) for
ordered crystals, typical for disordered solids can be seen in fig. 4. It makes solid solutions of the
investigated compounds preferable as potential moderators for cold neutron sources.
 J.Kalus, M.Monkenbusch, I.Natkaniec, M.Prager, JWolfrum, F.Worlen, Mol. Cryst. Liq. Cryst., 268 (1995) 1-20.
 C.Cavagnat, J.Lascombe, J.C.Lassegues, A.J.Horsewill, A.Heideman, J.B. Suck, J. Physique 45 (1984) 97-105.
 I.Natkaniec, K.Holderna-Natkaniec, J.Kalus, V.D.Khavryutchenko, in: Neutrons and Numerical Methods,
Grenoble 1998, Ed. M.R. Johnson, G.J. Kerlay, H.G. Buttner, AIP Proc. 479, p. 191-194.
 M.Utsuro, M.Sugimoto, J. Nuc. Sci. Technol., 14(5) (1977) 390-392.
 I.Natkaniec, K.Holderna-Natkaniec, in: Proc. of 6th Meeting of the Collaboration on Advanced Cold Moderators,
Juelich, 11 – 13 September 2002, Report FZ Juelich 2003.
PARTIAL STRUCTURE FEATURES OF PB−K MELT
N.M. Blagoveshchenskii1, V.A. Morozov1, A.G. Novikov1, V.V. Savostin1, A.L.
Shimkevich2, I.Yu. Shimkevich1
State Scientific Center RF, Institute for Physics and Power Engineering, 249033
Institute of Nuclear Reactors, Russian Scientific Center «Kurchatovsky Institute»
123182 Moscow, Russia
The paper is devoted to the extraction of the partial structure factors (PSF) and
partial pair correlation functions gij(r) (PPCF) from the results of the neutron
diffraction experiments. For this purpose we use our earlier experimental results on
Pb-K melt with various concentrations of components CPb, K ( fig. 1, see ).
The structure factor (SF) for Pb-K binary system can be written as:
S (Q) = (C Pb bPb ) 2 S PbPb (Q) + 2C Pb C K bPb bK S PbK (Q) + (C K bK ) 2 S KK (Q) (1)
Here bPb and bK are amplitudes of coherent neutron scattering. In coordinate space
(2π ) 2 n0 r ∫
g ij (r ) = 1 + [ S ij (Q) − 1] Q sin(Qr )dQ (2)
where n0 is density. To minimize the influence of spurious oscillations at small r
arising in the course of the procedure (2) two methods were used: the first, based on
the experimental SF correction , and the second, based on the maximum entropy
The PSF and PPCF, extracted from experiment, were analyzed by the
comparison with MD simulations results . In so doing two model of the melt
investigated were applied. In the first one the Pb-K melt was assumed as the mixture
of free Pb and K atoms (atom-atom (AA) approach). The example of PPCF gij(r) for
the melt with the concentration Pb0.86K0.14 obtained in the frame of this model, is
shown in fig 2. The second model assumes the melt to be a mixture of Pb-K clusters
and free Pb atoms (cluster-atom (CA) approach). It follows from MD results , that
Pb-K melt along with the clusters of Zintle type Pb4K4 (their existence is evident from
the presence of the prepeak in SF S(Q) at Q ∼ 1 A, , see fig. 1) contains as well the
Pb-K clusters with variations of components PbmKn . So, besides the Zintle clusters
the possible variants of the Pb-K clusters Pb2K2, Pb4K3, Pb4K2, and Pb6K4 were
considered, the latters being assumed as scattering units with coherent amplitude b =
mbPb + nbK. The examples of the experimental gij(r) for this AC approach are shown
in fig. 3 (g12(r) – PPCF for free Pb atom – Pb-K cluster correlations) and fig. 4 (g22(r)
– PPCF for Pb-K – Pb-K cluster correlations). As in the case of fig. 2, there exists
only far resemblance between MD and experimental results. It is desirable to explain
the possible origin of the main features demonstrated in these figures. The peak at r ∼
7 A may be connected with the correlations “cluster – free Pb”. The peak at (7 – 8) A
reflects the correlations between Zintle clusters. The peaks at ∼ (2-2.5) A, which are
absent in MD results can be understood presumably in the following way. Two Pb
atoms and two K atoms are situated at the vertices of tetrahedra. In this case the
shortest distance between scattering centers is 2 times less than interatomic
distance, i.e. 3.4/1.4 ≈ 2.4 Å. If the clusters converge through two neibouring free Pb
atoms, this distance can be estimated as twice of this value, i.e. ≈ 5 Å.
In general, one can point out some resemblance only between the positions of
the main features in MD data and PPCF curves extracted from experimental results, to
say nothing of their shapes.
 N.M. Blagoveshchenskii, Yu.V. Lisichkin, V.A. Morozov, A.G. Novikov, V.V.
Savostin, A.L. Shimkevich, I.Yu. Shimkevich. Appl. Phys. A74 (2002) 1107.
 D.K. Belashchenko. Crystallography (Rus) 43 (1998) 400.
 M.A. Howe, R.L. McGreevy, W.S. Howells. J. Phys.: Condens. Matter. 1 (1989)
 G.A. de Wijs, G. Pastore, A. Selloni, W. Van der Lugt. J. Chem. Phys. 103 (1995)
 H.T.J. Reijers, W. Van der Lugt, C. Van Dijk, M.-L.Saboungi. J. Phys.: Condens.
Matter. 1 (1989) 5229.
0.0 0 5 10 15 20 25
0 1 2 3 4 5 6 7 8 o
Fig. 1. Structure factor for the Pb−K melts for Fig. 2. PPCF g12(r) for the Pb0.86K0.14 melt:
various concentrations. 1 − MD result, 2 – AA approach.
3.0 2 1,4 2
0 5 10 15 20 25 0 5 10 15 20 25
Fig. 3. PPCF g12(r) for the Pb2K2 cluster: Fig. 4. PPCF g22(r): 1 – for Pb2K2 cluster,
1 – AC approach, 2 – MD result. 2 – for Pb4K4 cluster, 3 – MD result.
FIRST PHYSICAL RESULTS FROM REMUR
V.L.Aksenov, K.N.Zhernenkov, Yu.V.Nikitenko, A.V.Petrenko
Frank Laboratory of Neutron Physics, JINR, 141980 Dubna Moscow Reg., Russia
This year the first modernization stage of the polarized neutron spectrometer REMUR
completed. Today, the spectrometer allows carrying out high-luminosity investigations of neutron
reflection from surfaces, layered magnetic structures, and interfaces and experiments of small angle
neutron scattering on inhomogeneous magnetics for a wide interval of momentum transfer, Q =
3×10-3 ÷ 5×10-1Å-1.
The proximity effect at the superconductor-magnetic interface involving simultaneous
establishment of the superconducting and the magnetic state in the bilayer or periodic structure has
already been studied for a comparatively long time. In 1988, A.I.Buzdin and L.N.Bulaevskii 
predicted the effect of modification by superconductivity of the ferromagnetic order. Actually, it
was noted that in a thin ferromagnetic film a domain structure established. Contrary to  article 
points to that superconductivity leads to the establishment of a modulated magnetic structure.
Experimental studies of the effect of superconductivity on the ferromagnetic layer magnetization
were first conducted by Mühge and co-authors who determined the effective magnetization in a
Nb/Fe crystalline layer by measuring the magnetic resonance. They discovered that the
magnetization fell as the temperature decreased below the critical temperature and the thinner the
layer the sharper the fall was. For a thinnest iron layer of 14Å the magnetization drop was the
largest and amounted to 4%.
To reveal how superconductivity and magnetism co-exist on a nano-level, we have chosen
the layered structure Pd(15Å)/V(400Å)/Fe0.66V0.34(50Å)/ [10×(V(50Å)/Fe(50Å))]/MgO where
simultaneously exists the periodic structure 10×[V(50Å)/Fe(50Å)], that is composed of
superconducting vanadium and ferromagnetic iron layers, and the bilayer V(400Å)/
Fe0.66V0.34(50Å). The measurement at different temperatures of the dependence of the neutron
reflection coefficients R++(Q) and R--(Q) responsible for the processes without spin-flip has allowed
the determination of the spatial dependence of the magnetization (magnetization profile) in the
periodic structure and at the interface in the bilayer. At the same time, the given periodic structure is
a generator of short-period standing neutron waves , which also allows study of spatial variation
of the magnetization vector direction using the neutron reflection coefficients R+-(Q) and R-+(Q)
responsible for the neutron spin-flip processes.
Nb ,10 A 400Å
Nb ,10 A
6,6 Fe0.66/V0.3 4
81/7 88/10 100/14 90/10 83/9 15/7 10/10 0/14 8/10 15/9 0
90/10 M, [kOe] 50Å 400Å
М, [kOe] 100Å
93/7 93/9 98/7 100/10 100/14 98/10 78/9 40/10 38/10
Fig.1. The nuclear Nb and the magnetic profile M at 3К for: а) three contiguous layers
V(50Å) /Fe(50Å) /V(50 Å) of the periodic layered structure; b) bilayer Fe0.66V0.34(50Å)/V(400 Å ).
Measurements of neutron reflection were carried out at 293К, 7К, 3К, and 1.7К. Basing on
the fact that the temperature of superconducting transition in bulk vanadium is known to be 5.3К
and that it slightly decreases if in a nano-layer, it is assumed that the temperatures 3К and 1.7К lie
below the superconducting transition temperature for vanadium layers and the temperatures 7К and
293К are above it. However, the experimental data on neutron reflection coincide for the
temperatures 293К, 7К, and 3К and only differ for 1.7К.
Figure 1 shows the histograms of the spatial dependence of the nuclear scattering density
amplitude Nb(nuclear profile) and of the magnetization M (magnetic profile) for the three
contiguous layers V/Fe/V of the periodic structure (Fig. 1а) and the bilayer
Fe0.66V0.34(50Å)/V(400Å) (Fig. 1b) at 3К. In the Figure each separate section (structure sublayer) of
the histogram Nb is marked with the CFe and L values with a slash between standing for the iron
percentage in the sublayer CFe and the sublayer thickness L. It is seen that there are just two
sublayers that have 100% iron and 100% vanadium content (CFe=0). The rest are a mixture of iron
and vanadium atoms. Next, from a comparison of the dependence Nb with the dependence M it is
seen that in sublayers with a low iron concentration (15%, 10% or 8%) magnetization has a much
lower value than that which follows from the assumption of magnetization being proportional to the
iron atom concentration. The latter is due to the fact that in such sublayers vanadium atoms are
magnetized by iron atoms and get antiferomagnetically ordered with respect to them. For the bilayer
(Fig. 1b), we also have a 66% iron atom concentration in a 50Å thickness sublayer while the
magnetization is only 35% of the iron atom magnetization, which means that antiferromagnetic
ordering is stronger than in the periodic structure sublayers. Thus, in a real Fe/V layered structure
we have a more complicated case when together with ferromagnetic ordering in the middle of the
iron layer there exists antiferromagnetic ordering at iron-vanadium interface.
Figures 2a,b show the magnetic profile for the temperature 1.7К. In the Figure each sublayer
is marked with effective manetization Meff and L values with a slash between their. The Meff is
expressed in percentages the ratio of sublayer magnetization M and magnetization of iron atoms in
sublayer CFeMFe , where MFe is magnetization of iron equal 21.6kOe. It is seen that in the periodic
structure sublayers with a low iron atom concentration, 15%, 10%, 8% and 0%, magnetization
drops by 1кОе, 0.9кОе, 0.7кОе and 0, respectively. In the sublayers with a high iron atom
concentration, 81%, 88%, 100%, 90% and 83%, magnetization changes by –3кОе, -2кОе, 0, 0,
+2кОе, respectively. Thus, there is direct evidence of that magnetization tends to decrease, although
in separate sublayers it does not change and even increases. It is, however, characteristic of the
sublayer with a 100% vanadium atom concentration that in it there is not observed diamagnetism
due to establishment of superconductivity. At the same time, in the bilayer (Fig. 2b) the sublayer
Fe0.66V0.34 loses magnetization completely while the 400 Å vanadium layer becomes diamagnetic.
100Å 50Å M, [kOe] 50Å 400Å
60/7 60/9 80/7 89/7 100/14 98/10 89/9 \ -0,7
Fig. 2. The magnetic profile at 1.7К for: а) three contiguous levels V(50Å) /Fe(50Å) /V(50 Å) of the periodic layered
structure; b) bilayer Fe0.66V0.34(50Å)/V(400 Å ).
The results of the first neutron investigations are unexpected to some extent, though they
can be explained. Unexpected is that in the periodic structure with thin vanadium layers in each of
which superconductivity cannot occur  (what is seen on the example of the 14Å layer with a
100% vanadium concentration) changes in the magnetic profile are observed. A possible
explanation is that there establishes some superconducting state in the entire periodic structure
whose sum thickness of vanadium layers is on the order of 500 Å. Unexpected is 100% suppression
of a 5кОе magnetization in the Fe0.66V0.34 sublayer It may be connected with strong
antiferromagnetic ordering that produces a weaker destructive effect on the superconducting pair
So, the obtained results make us sure that the unique possibilities for the measurement of the
nuclear and the magnetic profile of layered structures that provide the present polarized neutron
spectrometer REMUR will allow us to obtain new data to advance essentially in the solution of the
problem of superconductivity-magnetism coexistence.
1. A.I. Buzdin, L.N. Bulaevsky, Sov. Phys. JETP 94 (1988) 256.
2. F.S. Bergeret, K.B. Efetov, A.I. Larkin, Physical Review B 62 (2000)11872.
3. Th. Mühge, N.N. Garifyanov, Yu.V. Goryunov et al., Physica C 296 (1998) 325.
4. V.L. Aksenov, Yu.V. Nikitenko, Physica B 267-268 (1999) 313; Physica B 297 (2001)101.
5. I.A. Garifullin, JMMM 240 (2002) 571.
6. C.L. Chien, D.H. Reich, JMMM 200 (1999) 83.
Nature of the parity violation in interaction of neutrons with lead
J.Andrzejewski 1), N.A.Gundorin 2), I.L.Karpikhin 3), L.Lason 1), G.A.Lobov 3), D.V.Matveev 2),
L.B.Pikelner 2), K.V.Zhdanova 2)
1) Lodz University, Lodz, Poland
2) Frank Laboratory of Neutron, JINR, Dubna, Russia
3) Institute of Theoretical and Experimental physics, Moscow, Russia
At the beginning 1980’s the enhanced parity nonconservation (PNC) effects were predicted [1, 2, 3]
and experimentally observed [4, 5] in the processes of interaction of slow neutrons with nuclei. It was
shown the structure and properties of the nuclei cause the mechanism of this enhancement. Those effects
are mostly expressed close to the p-wave resonances. For example, for 139La total capture cross section of
the p-resonance with the energy 0.75 eV differs by 10% for polarized and unpolarized neutrons.
Later on detailed investigation of these effects for a number of nuclei was carried out in Los-Alamos
, where the dependence of the neutron total cross section vs. neutron helicity was measured. All the
results are in agreement with the theory treating PNC-effects as a result of mixing of compound states,
having different parity. In the considered case they are s- and p- resonance.
Besides PNC-effect at total cross-section there is another effect - neutron spin rotation, if neutron
polarization is perpendicular to the neutron momentum, by neutron flight through the target. Both effects
are described in the frameworks of the same theoretical model.
Lead is among those nuclei where neutron spin rotation was measured. The value of rotation angle
was obtained in :
∆ϕ = (2.24±0.33) · 10-6 rad/cm
The target was a natural lead, which consists of four isotopes. Additional experiment on natural lead
 confirmed that the effect is present. The obtained value in this experiment was
∆ϕ = (3.53±0.79) · 10-6 rad/cm.
The carried out measurement on an isotope 207Pb, which in a natural mix of 22 %, has shown, that
this isotope does not respond to this effect .
Further measurement was done with an isotope 204Pb , which in natural lead only 1.4 %, and the
next value of rotation angle was obtained
∆ϕ = (8±2) · 10-5 rad/cm
It was somewhat less than needed for interpretation of the effect but approximately, nevertheless,
could explain it.
In the frameworks of the simplified two-level model of the s- and p- resonance mixing spin-rotation
angle may be written as follows :
4πD 2 (1eV ) ρWsp Γns (1eV )Γnp (1eV )
( E − Es )( E − E p )
Here: D - is the neutron wave length, ρ - is the number of nuclei in cm3 of target, Wsp - is the matrix
element of the mixing by the weak interaction between the states with different parity, Γns и Γnp – neutron
widths of s- and p- resonances, Es и Ep – energy means of this resonances. Symbol (1eV) demonstrates
that this value is reduced to 1 eV of energy.
It is assumed in Eq. (1) that the total widths Γs и Γp << ( E - Es ) and ( E - Ep ), respectively.
Using known values of the parameters of s- and p- resonances of 204Pb , one may see that the
theoretical estimation of ∆ϕ is several orders less than was experimentally obtained. It is possible, that the
compound state corresponding to the p-resonance lies below the binding energy (so-called negative
resonance). It is obvious from Eq. (1) that the effect in thermal region at E < 0.1 eV is proportion to
√ Γns / Es, and √Γnp / Ep. Taking maximal value of this ratio ( Es = -3 keV и Γns (1eV) = 1.3 eV )  and
assuming also that for p-resonance the average width is Γnp (1eV) = 3 · 10-7 eV and Ep = D/10 =100 eV
(where D is the average interval between nuclear levels which is about 1 keV for 204Pb) one obtains ∆ϕ =
9 · 10-7 rad/cm, i.e. 2 orders less than obtained in the experiment. Here we used Wsp = 5 · 10-3 eV which is
slightly higher in comparison with the average value. The larger effect may be obtained, if Γnp would be
essentially larger and Ep smaller, respectively. For instance, if an increase Γnp by an order of magnitude
and arranging the resonance below binding energy by 5 eV, we obtained ∆ϕ = 6 · 10-5 rad/cm. It becomes
nearer to the experimental value on 204Pb, while still not enough to explain the measurement on the
natural lead. Thus interpretation of the parity violation in lead may be related with the strong p-resonance
close to the binding energy. Therefore, observation of the “negative” neutron resonance is a principal
The purpose of experiment and estimation of expected results
We have proposed to investigate the energy dependence of the neutron capture cross-section σγ (E)
and such way to carry out search of subthreshold resonance. It is known that σγ (E) is described by the
Breit-Wigner relationship. At low neutron energy E << E0 and Г << E0 it may be rewritten as
πD 2 (1eV )Γn0 Γγ
σ γs ( E ) = 2
ES E (2)
in the case of s-wave interaction, and as
πD 2 (1eV )Γn Γγ
σ γ (E) =
in the case of p-wave resonance.
In expressions (2) and (3) Гn0 and Гn1 are the reduced widths of s- and p- resonances, which do not
depend on the energy of neutrons. An important role in the p-wave cross-section plays centrifugal factor:
( kR ) 2
1 + (kR ) 2 (4)
Here: k = 1 / D – is the neutron wave number and R – is the radius of a nucleus. For lead the number
of V1 = 3 · 10-6 E. It is seen from here that the energy dependencies of the neutron cross-sections for s-
and p- waves are different:
σ γs ( E ) ≈ 1 / E σ γp ( E ) ≈ E
It is seen that the p-wave contribution (or its upper limit) may be obtained from the measurements of
the energy dependence of the radiation neutron cross-section from thermal energy up to 1-3 eV.
The technique of experiment and results
The experiment carried out on the neutron beam of the pulsed reactor IBR-2 of FLNP JINR in
Dubna. The well fulfilled technique of time-of-flight for spectrometry of neutrons was used. As follows
from the experimental results above [8,9] the negative resonance most probable could be observed in
radiation neutron capture of 204Pb. Thus, we used as a target lead enriched by an isotope 204Pb. It is
convenient to measure gamma-ray spectra from two targets simultaneously to decrease various systematic
uncertainties during the experiment. The first investigated target is enriched by isotope 204Pb and the
second one is a reference target having known 1/√E energy dependence of the capture cross-section.
As soon as an overlap of the gamma-peak from two components of composite target is absent it may
be possible to compare the squares under these peaks. So, on the base of this comparison an unambiguous
conclusion concerning the deviation of the energy dependence of the neutron radiation cross-section from
the 1/√E can be maid. At the initial stage of experiment as a reference target served copper, and further as
reference was the isotope 207Pb. This isotope contained in the investigated sample and the lines
appropriate to it were intensive enough in a measured gamma-spectrum.
Gamma-ray spectroscopy of radiative neutron capture carried out with the Combined Correlation
Spectrometer (COCOS). Its peculiarity consists of the combination of lonely semi-conductor detector
(HPGe) with high energy resolution and several scintillation gamma-detectors (BGO) with high
registration efficiency. The possibility of registration of two and more coincidence gamma quantums and
the correlation analysis of multi-parameter experimental data allows essentially to suppress a background
and to improve a ratio “peak-background” in the measured spectrum. The high-energy part of the
spectrum from 5 to 8 MeV was analyzed in the carried out experiment. Here, the probability of formation
of electron-positron pairs due to gamma-ray interaction with a germanium crystal prevails the probability
of a photoeffect and Compton-effect.
This part of spectrum is presented in figure 1. It is a gamma-ray spectrum from radiation capture of
neutrons with energy 0.04 eV, registered by HPGe detector in coincidence with BGO registration of the
photon with energy 511 keV, accompanying of positron annihilation. The peaks marked on spectrum as S
and D, single and double escape, respectively, correspond to the direct transition from the top exited
levels on the ground levels for compound-nucleus 205Pb, 208Pb and 64Cu.
Result of experimental data of processing several series of measurements with the enriched isotope
Pb total duration 575 hours shows, that the ratio of direct transitions intensities of target components
(205Pb/208Pb) does not grow with increase of neutron energy, as it was expected, but falls. The similar
energy dependence of analogous ratio (64Cu/208Pb) was observed in the control measurement with the
enriched sample on an isotope 207Pb and copper as the reference.
204 205 207 208 207 208
Pb ( n , γ ) Pb Pb ( n , γ ) Pb Pb ( n , γ ) Pb 63 64
C u ( n ,γ ) C u
[ 5707 ke V (D )] [ 6346 ke V (D )] [ 6 8 5 7 k e V ( S) ] [ 6894 ke V (D )]
300 204 205 63 64
Pb ( n , γ ) Pb C u ( n ,γ ) C u
[ 6 2 1 8 k e V ( S) ] [ 7 4 0 5 k e V ( S) ]
Counts / c h.
H PG e ( En = 0 .0 4 e V ) ;( γ − γ -c o in c .) ;t m e a s. = 1 5 6 h
5600 5800 6000 6200 6400 6600 6800 7000 7200 7400 7600 7800
G a m m a e n e rg y , k e V
Fig.1. High-energy part of the gamma-ray spectrum of radiative capture of neutron with
energy 0.04 eV for an investigated sample.
The experimental data of measurements are presented in figure 2. There are inverse relations
K( Pb/205Pb) and K (208Pb/64Cu) for six group of neutron energy. It is obtained after normalization on
the average meaning of these relations for two group neutron energy 8 and 20 meV, where the
contribution of a p-wave is negligible. This data are supplemented with the calculated result of neutron
capture cross-section ratio σγ(207Pb)/ σγ(204Pb) as function of neutron energy with the presumable negative
resonance parameters satisfying observed effect of parity nonconservation in lead.
2 ,5 2 ,5
2 ,0 2 ,0
1 ,5 1 ,5
1 ,0 1 ,0
0 ,0 1 0 ,1 1 5
N e utro n e ne rg y , e V
Fig.2. Neutron energy dependence of experimental meanings K. (×) - Pb/205Pb, (•) -
Pb/ Cu, curved line – calculated result.
The experimental data of the measurements shown that neutron capture cross-section ratio
σγ( Pb)/ σγ(204Pb) increases with the incident neutron energy. This is an indication of the existence of the
p-wave capture at an isotope 207Pb. Thus, the result of the carried out experiment gives a possibility to
conclude that isotope 207Pb have a “negative” resonance, which could explain parity non-conservation
effect in the natural lead. So, it seems very interesting the more detailed study of the nature of the
observed PNS-effect in lead.
The work was supported by RFBR (grants 01-02-16024 and 02-02-16935) and INTAS (project 00-
1. В.А. Карманов, Г.А. Лобов Письма в ЖЭТФ, 1969, т. 10, стр. 332
2. О.П. Сушков, В.В. Фламбаум Письма в ЖЭТФ, 1980, т. 32, стр. 377
3. V.E. Bunakov, V.P. Gudkov Zs. Phys., 1981, Bd. 303, s. 285
4. M. Forte, B.R. Heckel et al. Phys. Rev. Lett. , 1980, v. 45, p. 2088
5. В.П. Алфименков, С.Б. Борзаков и др. Письма в ЖЭТФ, 1981, т. 34,
стр. 308; 1982, т. 35, стр. 42
6. G.E. Mitchell et al. Physics Reports, 2001, v. 354, p. 157241
7. B.R. Heckel, N.F. Ramsey et al. Phys. Lett. B, 1982, v. 119, p. 298
8. В.П. Болотский, О.Н. Ермаков и др. ЯФ, 1996, т. 59, стр. 1873
9. R. Golub, I.L. Karpikhin et al. Proc. IX Intern. Seminar on Interaction of Neutron with
Nuclei, Dubna, 2001, p. 33
10. V.E. Bunakov and L.B. Pikelner Progr. Part. Nucl. Phys., 1997, v. 39, p. 337
11. S.F. Mughabghab Neutron Cross Section, 1984, V. 1, Part B
Development of Neutron Polarizer-Analyzer System for T-Invariance Experiment
V.R. SkoyA1, Y. MasudaA2, S. MutoA3, T. InoA3, G.N. KimA4
Frank Laboratory of Neutron Physics, Joint Institute for Nuclear Research, 141980 Dubna,
Moscow Region, Russia
Institute of Particle And Nuclear Studies, High Energy Accelerator Research Organization,
Tsukuba, Ibaraki, 305-0801, Japan
Institute of Materials Structure Science, High Energy Accelerator Research Organization,
Tsukuba, Ibaraki, 305-0801, Japan
Institute of High Energy Physics, Kyungpook National University, 1370 Sankyok-dong, Buk-gu,
Daegu 702-701, Korea
A polarized 3He neutron spin filters are useful for numerous researches in fundamental
physics and applied physics with neutrons . We will apply the polarized 3He neutron spin filter
for testing time reversal invariance in the nuclear reaction. The present experiment is a
methodological preparation for the T-violation experimental, namely R&D for the experimental
apparatus which will be used in the forthcoming full-scale experiment. The experiments were
performed on KENS neutron beamline H-8.
Any noble gas nucleus of odd isotope is highly polarized by means of a rubidium optical
pumping . In the optical pumping, circularly polarized photons of 795 nm in the wavelength are
absorbed at the rubidium D1 resonance. Upon the resonance absorptions, photon circular
polarizations are transferred to rubidium atomic electrons, and then rubidium atomic spins are
polarized in the direction of photons. Polarized rubidium atoms collide with noble gas nuclei of
non-zero spins. Upon the atomic collisions, the rubidium atomic polarization is transferred to the
He nuclear polarization via hyperfine interactions.
Neutrons are polarized upon transmission through the polarized 3He gas by the optical
pumping because of strong spin-dependent neutron absorption by the 3He(n,p)3H reaction. The
neutron polarization is represented by the following expression:
pn = tanh ( p He nHe σ P L ) , (1)
where nHe is a He nuclear number density, p He is a He polarization, and L is a target length.
Quantity σ p is a polarization cross-section of the 3He(n,p)3H reaction, which is a difference
between cross sections for the parallel and antiparallel spin states of a neutron and 3He nucleus
system. The polarization cross section depends on the neutron energy, E as
σ p (E ) ≈ 5370 0.0253 E barn. The neutron transmission is represented as
N = exp(− nHe σ 0 L ) cosh ( p He nHe σ P L ) = N 0 cosh ( pHe nHe σ P L ) , (2)
where N0 is a neutron transmission for unpolarized He nuclei and σ 0 is the total cross-section of
the 3He nucleus. From Eq. (1) and (2), the neutron polarization is obtained as
pn = 1 − ( N 0 N )
In the test of the time reversal invariance, polarized 3He neutron spin filters are used as a
neutron spin polarizer and an analyzer. A prototype set-up for the polarizer and analyzer are shown
in Fig. 1. Two 3He cells are placed in two solenoids of 34 G as the polarizer and analyzer. Each cell
is irradiated with a laser beam of 795 nm in wavelength. A neutron beam passes through the
polarizer and analyzer in the axis of the solenoids. The configuration of the apparatuses can be
changed easy and fast enough to fit the particular experiment. In fact, it represents the constructor
set of uniform parts. For example, we can place a spin flipper for a neutron spin manipulation
experiment as it is shown in Fig. 1. The solenoids are placed on revolving platform, which allows
rotate the entire installation around the upright axis and sets the angular position with accuracy
Fig. 1. Installation with the polarized 3He cells in a KENS beam line. Position of the neutron
detector is not scaled.
As a container for polarized 3He gas, we used a 3 cm diam and 4.65 cm length cylindrical
cell of quartz and a 3.6 cm diam and 4.65 cm length cell of sapphire. The both cells were filled with
He gas at a pressure of 3 atm. together with small amount of rubidium and nitrogen gas. The cells
were placed in the two solenoids aligned along the neutron beam direction. The quartz and sapphire
cells have the advantage of clean surface for polarized 3He gas and low neutron absorption. Each
cell had its own hot air supply line. The temperature of the cells was kept at about 195±1oC so that
an optimal rubidium atomic number density was obtained for the optical pumping.
Fig. 2. LDA FWHM narrowing system. Mirror M1 guides zeroth-order reflection from grating to
the mirror M2 which reflects the light to a cell inside the solenoid. All lenses are used for shaping of
the laser beam profile.
For the optical pumping, we used 17 W laser diode arrays (LDA) of 795 nm in wavelength
as shown in Fig. 2. The LDA has a line width of 2~3 nm in FWHM and are much broader than the
Rb natural absorption line width. Therefore, a small fraction of the LDA power can be used for the
rubidium optical pumping. The effective laser power limits the rubidium atomic polarization and
then the 3He nuclear polarization. There are two ways to improve this situation. First one is a brute
force, just to increase the laser power. We used two laser systems for single cell. We used almost
twice as much laser power. The second one is the application of a frequency narrowing system to
the LDA, which allows us to use full laser power . We used an external cavity with a diffraction
grating and some optical lenses for the frequency narrowing. The grating has 2400 grooves/mm.
The first order diffraction from the grating is reflected back to the LDA and then form a cavity in
order to stimulate coherent photon emissions. The 0th-order diffraction is used for the extraction of
the laser beam. The result of the frequency narrowing is shown in Fig. 3. As shown in Fig. 3, the
width of the LDA is reduced to the rubidium D1 resonance width. The extracted laser beam is
guided through an optical lens system for a beam shaping for the 3He cell irradiation.
Fig. 3. The result of the frequency narrowing system.
The transmission enhancement, N 0 N was measured by a neutron detector of 10B-loaded
liquid scintillator, which is placed at 12 m from the pulsed neutron source, as a function of a
neutron time of flight. From Eq. (3), the neutron polarization was obtained. The difference of
incident neutron intensities between N0 and N measurements was normalized by using the property
of the polarization cross section. The polarization cross section is almost zero, and then the
transmission enhancement, N 0 N is almost unit at higher neutron energies above 120 eV. The
deviation from unity is less than 0.01. The results of neutron polarization for the quartz and sapphire
cells are shown in Fig. 4 and Fig. 5.
PHe = 0.538 ± 0.004stat ± 0.02sys PHe = 0.627 ± 0.002stat ± 0.01sys
0.01 0.1 1 10
0.01 0.1 1 10
Fig. 4 Neutron polarization with quartz cell. Fig. 5 Neutron polarization with sapphire cell.
A 3He pressure in each cell was extracted from neutron transmissions through each cell and
an identical empty cell, and then the 3He nuclear number density was obtained. The 3He
polarization was determined by means of Eq. (1). The result of the 3He polarizations are pHe
= 0.538 ± 0.004 stat ± 0.02 sys for the quartz cell and pHe = 0.627 ± 0.002 stat ± 0.01sys for the sapphire
cell, respectively. In the both cases the single laser with the narrowing system was applied to the
LDA. The double laser pumping for the same quartz cell without the narrowing system provided pHe
= 0.431 ± 0.002 stat ± 0.02 sys. The error of the 3He polarization in the quartz cell is bigger than in the
sapphire cell. The increase in the error arises from the uncertainty of 3He pressure. The 3He gas
diffuses through the cell wall. The 3He pressure decreased at a rate, ∼1% per day at 195oC. The
leakage was not found for the sapphire cell.
We have compared different methods of optical pumping for the 3He polarization. The
application of the frequency narrowing system to the LDA showed better performance than the
direct increase in the laser power. The sapphire cell showed better performance in the 3He
polarization than the quartz cell, in addition, the sapphire cell had no 3He gas leakage.
1. Proceedings of “HELION97 from Quark to Life Workshop” (1998) Nucl. Instr. Meth. A402,
2. Cates G.D., Shaefer S.R. and Happer W. (1988), Phys. Rev., A 37, 2877.
3. Chann B., Nelson I. and Walker T. G. (2000) “Frequency-narrowed external-cavity diode-
laser-array bar”, Opt. Lett., 25, 1352-1354.
STELLAR NEUTRON CAPTURE ON PROMETHIUM: IMPLICATIONS FOR
THE s-PROCESS NEUTRON DENSITY
R. Reifarth, C. Arlandini, M. Heil, F. Käppeler
Forschungszentrum Karlsruhe, Institute für Kernphysik, Karlsruhe, Germany
P. V. Sedyshev
Frank Laboratory of Neutron Physics, Joint Institute for Nuclear Research, Dubna, Russia
Ente Nazionale per le Nuove Tecnologie, l’Energia, e l’Ambiente, Applied Physics Division,
International Atomic Energy Agency, Nuclear Data Section, Vienna, Austria
Departement für Physik und Astronomie, Universität Basel, Basel, Switzerland
Istituto di Fisica Generale, Universitá di Torino and Sezione INFN di Torino, Torino, Italy
Max-Planck Institut für Astrophysik, Garching, Germany
The unstable isotope 147Pm represents an important branch point in the s-process reaction
path (Fig. 1). The resulting abundance ratio of 148Sm and 150Sm is affected by three branchings at
the unstable isotopes 147Nd, 147Pm, 148Pm. In framework of the classical s-process, based on the
assumption of steady process with constant temperature and neutron density, the strength of
branching can be expressed in terms of the rate for β-decay and neutron capture of the branch
point nucleus as well as by the σ N s values of the involved isotopes,
λβ (σ N s )branched
fβ = ≈
λ β + λn (σ N s ) unbranched
The effective strength of the combined branchings at A = 147/148 is determined by the branched
and unbranched s-only isotopes 148Sm and 150Sm. These isotopes are shielded against β-decays
from the r-process, hence, σ N s = σ N . The isotopic ratio is well defined , the cross
section ratio 148Sm/150Sm was measured with high accuracy . Such a way, the effective
branching factor is known - f β = 0.870 ± 0.009. Since the β-decay rates of the branch point
nuclei 147Nd, 147Pm, 148Pm are practically independent of temperature and electron density during
s-process , the expression can be solved for λ n = nn σ v T , choosing the neutron density nn
such as to reproduce 148Sm and 150Sm in solar proportions. However, the uncertainties lie
evidently in the cross sections of the branch point isotopes, for which only theoretical values
existed so far.
The stellar (n,γ) cross section for 147Pm was determined via the activation technique. The
quasy-stellar neutron spectrum was obtained by bombarding a thick metallic Li target with
protons of 1912 keV at the Karlsruhe Van de Graaff accelerators (Fig. 2). The sample was a
graphite pellet 6 mm in diameter and 0.4 mm in thickness, containing a mixture of 147Pm and
Sm (as a carrier). The sample was sandwiched between gold foils and placed on the lithium
target. The experiment was difficult because the relatively short 147Pm half-life of 2.62 yr
enforced the sample mass to be restricted to 28 ng or 1014 atoms only. By means of a modular,
high-efficiency Ge Clover array (Fig. 3) the low induced activity could be identified in spite of
considerable backgrounds from various impurities (Figs 4 and 5). Both partial cross sections
feeding the 5.37 day ground state and the 41.3 day isomer in 148Pm were determined
independently, yielding a total (n,γ) cross section of 709 ± 100 mb at thermal energy of kT = 30
keV. This is clearly smaller and at lest a factor of 2 more accurate than the recommended
theoretical value (1290 ± 350 mb) .
The (n,γ) cross sections of the additional branch point isotopes 147Nd and 148Pm as well as
the effect of thermally excited states were obtained by detailed statistical model calculations.
Three different sets of calculations using the same Hauser-Feshbah statistical model but different
parameterizations have been considered: NON-SMOKER, EMPIRE-II, ENEA Code. The
calculations lead to different cross sections, however the ratios of these results, 148Pm/147Pm,
Pm/147Pm are consisted within 8%. In combination with the measured value of 147Pm, this
provides a reliable estimation of cross sections for 148Pm and 149Pm. It has a significant impact on
the theoretical assessment of another important branch point isotope 148Pm. The uncertainty of
this cross section could be reliable reduced to 15%.
The present results allowed considerably refined analyses of the s-process branchings at A
= 147/148. The classical analysis yields a neutron density nn = (4.94 +0..60 ) ×108 cm-3, what is in
− 0 50
conflict with recent studies of the branching at A = 191/192  (nn = (7 +0..5 ) ×107). This
inconsistency confirms once more that the s-process mechanism must include a dynamic
component. In agreement with observations, the main s-process component is commonly ascribed
to He shell burning in thermally pulsing stars on the asymptotic giant branch (TP-AGB stars)
with masses 1.5-3 M . During the interpulse period of a few times 104 yr the dominant
C(α,n)16O reaction provides a relatively high neutron exposure at comparably low temperature
(kT ≈ 8 keV) and neutron density (nn ≤ 107 cm-3). At the start of the next convective instability, a
sufficiently high temperature is reached at the bottom of the He-burning to marginally activate
the 22Ne(α,n)25Mg source. This short burst of ~5 yr reaches peak neutron density of nn ≤ 1010 cm-
. Although this second burst represents only a few percent of the total neutron exposure, it
suffices to determine the final abundance pattern of the s-process branchings. In this stellar
model, the neutron density of 13C(α,n) source is not sufficient for the reaction flow to bypass
Sm. Only in the second burst of 22Ne(α,n) source, 148Sm is bypassed and even strongly
depleted (Fig. 6). However, during the final decline of the neutron density, the branchings to
Sm are restored, and the final value is established during the freezeout of the abundance
pattern. With the new cross sections, these models reproduce the observed abundance ratio
Sm/150Sm to better than 1%. The effective parameters obtained by the classical analysis have to
be considered as local features. This emphasizes the importance of the decline of neutron density,
which leads to a stepwise freezeout of the abundance according to the particular cross section
situation in each branching.
1. E. Anders, N. Grevesse. Geochim. Cosmochim. Acta. 53 (1989) 197
2. K. Wisshak et. al. Phys. Rev C 48 (1993) 1401
3. K. Takahashi, K. Yokoi. At. Data Nucl. Data Tables. 36 (1987) 375
4. Z. Y. Bao et. al. At. Data Nucl. Data Tables. 76 (2000)
5. P. Koehler et. al. Nucl. Sci. Tech. 2 (2002) 546
6. F. Kāppeler et. al. Ap. J. 354 (1990) 630
Fig 1. s-process path between Nd and Sm
partly bypassing 148Sm as a result of the
branchings at A = 147/148. The second s-only
isotope, 150Sm, experiences the full reaction
flow. An additional, very small branching to
Pm has been omitted for better readability of
the figure but was considered in all analyses.
The half-lives reflect the stellar values.
Fig 2. Sketch of the activation setup at the
Van de Graaff accelarator. Fig 3. Schematic view of the HP Ge
spectrometer consisting of two fourfold
Clover-type detectors in close geometry.
Fig. 5. γ-ray spectrum of coincident events
from the activated 147Pm sample. The spectrum
was obtained by off-line analysis of the data
taken with the Ge Clover array. Events due to
cascades from the decay of 148gPm consisting
Fig 4. GEANT simulation of the decay of
148m of the 915 and 550 keV transitions are
Pm with the Clover system operated in
concentrated in the center, clearly separated
calorimetric mode. The complex response
from the overall background and from
illustrates the need for a detailed simulation,
Compton-scattered events of the 1461 keV line
since cascade corrections for summing-in and
from 40K, which appear as the diagonal band in
summing-out effects no longer be handled with
the lower left part.
codes for single Ge detectors.
Fig. 6. Evolution of neutron density (right
scale) and of the abundances of 148Sm and
Sm in fractions of mass during the 22Ne
neutron release in a typical advanced pulse
(pulse 15 of the standard AGB model). The
timescale starts at the moment when the
bottom temperature reaches 2.5×108 K. The
arrow indicates the freezeout of the 148Sm
abundance according to the criterion Xfreeze =
ACTIVE BIOMONITORING WITH MOSS-BAGS APPLIED TO
AN INDUSTRIAL SITE IN ROMANIA
O. A. Culicov1, R. Mocanu2, M.V. Frontasyeva1, L. Yurukova3, E. Steinnes4
Frank Laboratory of Neutron Physics, JINR, Dubna, Russia
Departament of Analytical Chemistry,“Al.I. Cuza” University, Iasi, Romania
Institute of Botany, Bulgarian Academy of Sciencies, Sofia, Bulgaria
Norwegian University of Science and Technology, Trondheim, Norway
The well known ability of mosses to sorb and retain elements from wet and dry
deposition (Clymo, 1963) was successfully used during the last 30 years to identify and
monitor zones of heavy metal contamination by the collection and analysis of moss samples
(Buse et al., 2003). Since the introduction of the method (Rühling and Tyler, 1968), the most
popular variant has been passive monitoring, involving only two major steps: collection and
analysis of moss samples. This method has the advantage of the extensive character but does
not offer information about exposure of the moss to the pollutant agent during periods of less
than a year. The active biomonitoring with transplanted bryophytes in moss-bags, introduced
by Goodman and Roberts (1971), has the distinct advantage of a well-defined exposure time
This work is the first application of the moss-bag technique in Romania. In order to
optimize the assessment of atmospheric pollution in an industrial area using active
biomonitoring a novel sampling design was introduced, and transplants with two different
qualities of the moss Sphagnum girgensohnii were deployed in parallel in order to study the
uptake of a series of trace elements from the air over a defined time period. The site selected
for this experiment was Baia Mare, Romania (47°44' N, 23°20' E, altitude 228 m),
characterized by a sub-Mediterranean climate (an exception for a such latitude) and high
pollution with heavy metals from non-ferrous ores mining and metallurgy (Figure 1).
Figure 1. Sampling points and place of transplant
Nine moss transplants from each of two background areas (Dubna, Russia and Vitosha
Mountain Natural Park, Bulgaria) were deployed in parallel on balconies about 24 m above
street level for 4 months (Figure 2).
Figure 2. T system used for deploying moss-bags
Conventional and epithermal neutron activation analyses at IBR-2 pulsed fast reactor
FLNP JINR Dubna, Russia (Frontasyeva and Pavlov, 2000), were used to determine the
contents of 36 elements in moss. The analytical quality control was ensured by carrying out
concurrent analyses of the standard reference materials.
A comparison of the two moss qualities used in the transplant experiment shows that the
moss from Russia (RU) was a better choice than those from Bulgaria (BG) because of
generally lower content of many elements of interest. Obviously the BG moss had a greater
content of soil particles than RU as evident from the markedly higher levels of elements such
as Al, Sc, Fe, REE, Ta, Th, and U. Moreover higher contents of V, Zn, As, Cd, and Sb in BG
may indicate some exposure of the growing site from the Sofia region, which is only about 25
km away from Vitosha Mountain. Nevertheless, the individuals of moss collected in Bulgaria
were drier and weaker than the moss in Russia, with a yellow-green color compared to green
of moss from Russia, indicating a worst development of the Sphagnum in Bulgaria than in
In order to study the relative uptake of the different elements relative to the initial levels
in the transplanted moss an "Increment coefficient" ICx is introduced:
Tx ⋅ (Tx − U x )
IC x =
U x⋅ S x
where: ICx = increment coefficient for the element X;
Tx = concetration of element X in transplanted moss;
Ux = concetration of element X in unexposed moss
Sx = uncertainty of ICx
The investigated elements may be grouped in three groups as follows:
1. ICx has a significant positive value (ICx >1), indicating a net uptake of element X relative
to the initial content.
2. ICx has a significant negative value (ICx<0), indicating a net loss relative to the original
content of X.
3. ICx is not significantly different from 0 (0<ICx<1).
Not surprisingly the group of elements with ICx >2 includes typical exponents of air
pollution in surroundings of Pb and Zn smelters (Se, As, Sb) and elements associated to wind
blown particles from dumps and ore transport by trucks. The next group (1<ICx<2) contains
elements more likely to be connected with a crustal component absorbed by the moss in the
form of windblown soil dust. The group with ICx not significantly different from 0 most
probably represents elements lost from the moss at about the same rate as they are supplied
from the atmosphere.
The way IC values come out for RU and BG transplants, respectively is a very strong
demonstration of the necessity to start any field work with a “clean” moss.
The elements subject to major loss from the moss during the deployment period are
physiologically active elements such as Cl and the alkali elements Rb and Cs, which
presumably exhibit a behavior similar to that of K in the moss. These elements were lost in
about the same proportion in RU and BG. A similar situation was noticed by Yurukova and
Ganeva (1997) for other two Sphagnum species, for Ca, Mg, Na and especially K. No
previous data on, Rb, and Cs in this respect were found in the literature. In the present case,
the mentioned elements were most probably removed from Sphagnum tissues due to a
combination of two factors: accumulation of heavy metals and desiccation/ hydration cycles.
When dry the Sphagnum moss becomes brittle and the small leaf fragments tend to fall
off the stem. In this way some of the leaf biomass may be lost and the final sample subjected
to analysis may have a different leaf/stem mass ratio than the original moss. It was therefore
considered necessary to determine this ratio in the employed mosses for the elements in
question. Samples of RU consisting of only leaf biomass were analyzed and compared to the
corresponding data for whole moss. The following trends are evident:
leaf content higher than whole moss: Mg, Al, Sc, Cr, Mn, Ni. Fe, Co, As, Se, Br, Rb, Mo,
Ag, Sb, I, Ba, REE, Hf, Ta, Au, Th, U;
no significant difference: Na, Cl, K, Ca, V, Mn, Zn, Sr, Cd, In, Cs, I, W.
The second group includes elements active in moss physiology and elements with
similar chemical behaviour. These elements are mobile and presumably present in all cells in
the moss. The first group on the other hand may have been supplied to the leaf surface and
fixed there, and is thus transferred to the stem only to a limited extent. Since this group
contains most of the elements of interest from the pollution monitoring point of view it is
important that the leaf/stem ratio is not significantly changed during the transplant
In conclusion, the moss-bags using Sphagnum girgensohnii demonstrate a good or very
good capacity of response to the environmental conditions for a majority of the 36
investigated elements. This capacity depends not only on the moss species as demonstrated in
some previous studies, but also on the initial state of the transplanted moss. The higher the
element concentrations are in the moss before exposure the lower are the values of the
increment coefficient defined in this work. Based on differences in element content between
leaves and whole moss it is clearly suggested that the moss samples used in a monitoring
campaign should retain a constant leaves-to-stem ratio over the exposure period. The success
of the active biomonitoring with moss-bags also depends on the instrumental method used to
determine the elements. This method should be able to detect concentrations not only in
exposed but also in unexposed moss with as small error as possible. NAA fully demonstrates
this capacity for a great number of elements.
Buse, A., Norris D., Harmens, H., Buker, P., Ashnden, T., Mills, G.: 2003, Heavy metals in
European mosses: 2000/2001 survey, UNECE ICP Vegetation.
Clymo, R.S.: 1963, `Ion exchange in Sphagnum and its relation to bog ecology`, Ann.Bot. 27,
Frontasyeva, M.V., Pavlov, S.S.: 2000, `Analytical investigations at the IBR-e reactor in
Dubna, Preprint, JINR, Dubna, E14-2000-177.
Goodman, G.T. and Roberts, T.M.: 1971, `Plants and soils as indicators of metals in the air`,
Nature 231, 287-292.
Rühling, Å. and Tyler, G.: 1968, `An ecological approach to the lead problem`, Bot Notiser.
Steinnes, E.: 1989, `Biomonitors of air pollution by heavy metals`, in: Pacyna, J.M. and Ottar,
B. (eds), Control and Fate of Atmospheric Trace Metals, Kluwer Academy Publishers,
Dortrecht, pp. 321-338.
Yurukova, L. and Ganeva, A.:1997, `Active biomonitoring of atmospheric element deposition
with Sphagnum species around a copper smelter in Bulgaria`, Angew. Bot. 71, 14-20.
NEUTRON ACTIVATION ANALYSIS FOR DEVELOPMENT
OF MERCURY SORBENT BASED ON BLUE-GREEN ALGA
L.M. Mosulishvili1, A.I. Belokobylsky1, A.I. Khizanishvili1,
M.V. Frontasyeva2, E.I. Kirkesali2, N.G. Aksenova2
E. Andronikashvili Institute of Physics of the Georgian Academy of Sciences,
Frank Laboratory of Neutron Physics, JINR, Dubna, Russia
Mercury and its compounds are widely used in various branches of industry,
agriculture and medicine penetrating the environment in one or another way. A considerable
anthropogenic part of the environmental pollution by Hg is contributed by Hg
pyrometallurgy, non-ferrous metallurgy, production of chlorine and caustic soda,
consumption of fuel, garbage etc.
Mercury holds the first position for toxicity among other heavy metals. Medico-
biological studies of the last decades showed the gravity of the «mercury hazard» related to
the transition of chronic poisoning by Hg vapor from the professional diseases into the disease
Thus, the necessity to study the peculiarities of Hg interaction with living systems is
obvious. A blue-green microalgae Spirulina platensis (S. platensis), which is widely used as a
basis for pharmaceuticals and also as a biologically active food additive for humans and
animals, is considered as a living system.
Algae are often used in water remediation from heavy metals [1–3]. The processes of
accumulation and adsorption of mercury by biomass of the blue-green alga S. platensis
depending on the Hg concentration in the medium, where the growth of spirulina cells occurs,
MATERIALS AND METHODS
Cultivation of S. platensis was carried out in a standard Zaroukh alkaline water-salt
medium, mercury glycinate (HgNCH2COOH) was used as a nutrient loading.
In the first series of experiments to study the Hg accumulation by the S. platensis cells
the concentrations of nutrient medium loading by mercury constituted 100, 50, 5, 1, 0.1
µg Hg/L. Samples in all the series were taken every 24 hours.
In the second short-term series of experiments to study the Hg adsorption by
S. platensis concentration of nutrient medium loading was 500 µg Hg/L. Dynamics of the
adsorption processes, usually taking place during 1-2 hours, were studied during 1 hour.
Samples were obtained in 2, 10, 20, 40 and 60 minutes after the beginning of cultivation.
Mercury content in the samples was determined by epithermal neutron activation
analysis (ENAA) at the pulsed fast reactor IBR-2 (FLNP JINR, Dubna). Earlier we used the
technique of ENA analysis of S. platensis samples both to determine its background elemental
content and to study accumulation processes of some trace elements [4,5]. The samples were
irradiated for 5 days and their activity was measured twice in 4 and 20 days. The mercury
content was determined by γ-line with the energy 279.1 keV of isotope 203Hg. Here the
influence of interference lines 75Se and 182Ta was taken into consideration. The ENAA data
processing and determination of Hg concentrations were performed with the help of programs
used in FLNP JINR.
RESULTS AND DISCUSSION
The results of experiments to study Hg accumulation from nutrient medium by the
Spirulina platensis biomass at cell cultivation during 6 days at various Hg concentrations are
presented in Fig.1. In all the cases the exponential character of decrease of Hg content is
observed. The curves are well approximated by the function y=y0+Ae-x/t.
Hg in dry Spirulina pl. biomass (ppm)
Hg in dry Spirulina pl. biomass (ppm)
Hg load - 100 µg/L
5 Hg load - 50 µg/L 2.5
Hg load - 5 µg/L
1 2 3 4 5 6 0 10 20 30 40 50 60
Day Time (min.)
Fig. 1. Hg accumulation from nutrient Fig.2 . Hg(II) adsorption by the Spirulina
medium by the Spirulina platensis biomass platensis cells.
at various loading during 6 days.
Such character of dependence seems to be clear, as the number of S. platensis cells
grows exponetially, the number of sites of Hg(II) ion binding surpasses considerably the
number of Hg(II) ions in nutrient medium. This results in blocking of toxic Hg ions and their
removal from the nutrient medium. Such mechanism may serve as one of the important ways
for biosphere to «self-purify» from heavy metals with the help of microorganisms.
The results of investigation of Hg adsorption process by the S. platensis cells are
presented in Fig. 2. The experimental data obtained by ENAA method approximate well by
the polynomial of the third order: y=0.3586-0.02286x+0.00332x2-0.0000406482x3. As seen
from the obtained curve, the maximum Hg content is adsorbed by the S. platensis biomass
within 50 minutes and then a diminution of concentration is observed. Similar character of
dependence of Hg(II) accumulation was also obtained in paper .
If we take into account that Hg content in control samples constituted approximately
0.007 ррm, than it turns out to be that in 50 minutes the S. platensis biomass accumulates
mercury in about 300 times more. Thus, at relatively low Hg concentrations (of the order of
100 µg/L) in the medium S. platensis can be used in the remediation of industrial and sewage
waters from mercury.
Here, it should be also noted, that the S. platensis biomass consisting of long trichoms
can be easily gathered (separated) by filtration, which makes the technological process
considerably cheaper and simpler.
1. By the ENAA method it is possible to control the rate of Hg assimilation from nutrient
medium by the S. platensis biomass in the course of its cultivation in open ponds.
2. At Hg concentrations of the order of 100 µg/L the S. platensis biomass in its natural state
may be used to accumulate Hg(II) ions for the purpose of their removal from the
3. The S. platensis biomass is suitable for fast remediation of industrial and sewage waters
from mercury by way of biosorption and subsequent separation with the help of filtration.
1. D. Schmitt, A. Műller, Z. Gsőgőr, F.H. Frimmel, C. Posten. The adsorption kinetics of metal ions
on to different microalgae and silicens earth. Wat. Res. 35, 3, 2001, 779-785.
2. D.W. Darnall, B. Green, M.T. Henzl, J.M. Hosea, R.H. Mc. Pherson, J. Sneddon, M.D.
Alexander. Recovery of Heavy Metals by Immobilized Algae. In R. Thompson (Ed.) Trace Metal
Removal from Aqueous Solution, Whitstable Lito Ltd., 1986, 1-24.
3. N. Rangsayatorn, E.S. Upathum, M. Kruatrachue, P. Pokethitiyok, G.R. Lanza. Phytoremediation
potential of Spirulina (Arthrospira) platensis: biosorbtion and toxicity studies of cadmium.
Environ. Pollut., 119, 2002, 45-53.
4. L.M. Mosulishvili, Ye. I. Kirkesali, A.I. Belokobilsky, A.I. Khizanishvili, M.V. Frontasyeva, S.F.
Gundorina, C.D. Oprea. Epithermal neutron activation analysis of blue-green algae Spirulina
platensis as a matrix for selenium-containing pharmaceuticals. J. Rad. and Nuc. Chem., 252, 1,
5. L.M. Mosulishvili, Ye. I. Kirkesali, A.I. Belokobilsky, A.I. Khizanishvili, M.V. Frontasyeva, S.S.
Pavlov, S.F. Gundorina. Experimental substantiation of the possibility of developing selenium-
and iodine-containing pharmaceuticals based on blue-green algae Spirulina Platensis. J. Pharm.
And Biomed. Analysis, 30, 2002, 87-97.
6. Y. Kaçar, G. Arpa, S. Taw, O. Denizli, Ö. Genc, M.Y. Azicc. Biosorption of Hg(II) and Cd(II)
from aqueous solutions: comparison of biosorptive capacity of alginate and immobilized live and
heat inactivated Phanerochaete chrysosporium. Process Biochemistry 37, 2002, 601-610.